Delper Group Executive Summary

1. The Global Problem & Macro Inefficiencies

• The Global Logistics Bottleneck: Total global logistics spending commands between $11 Trillion and $13 Trillion annually , crippled by fragmented infrastructure and structural delays.
• The $800 Billion Administrative Drain: Cross-border friction, manual customs compliance, and port demurrage waste up to $800 Billion in economic value each year.
• Physical Sourcing Vulnerabilities: Industrial supply chains face severe cyber threats on legacy Operational Technology (OT), combined with a global shortage of qualified cyber-physical engineering talent.
• Digital Commerce Vulnerabilities: Global B2B and retail e-commerce networks lose massive volumes to counterfeit goods, phantom inventory, and upfront capital exposure without independent quality validation.

↑ Index

2. The Integrated Solution: Four Initiatives, One Ledger

Delper Integrated merges four initiatives into a single economic engine, operating exclusively via the 100% fiat-backed ERC-20 Delper Token:

• Project Iron Nexus ($46.00 Trillion): A global multimodal transport grid linking deepwater ports, dedicated electrified rail corridors, express highways, and high-velocity air cargo hubs. Rail freight is designed to cut transport energy use by 60%.
• Project Make in India 🇮🇳 ($56.00 Trillion): A coastal software-defined manufacturing matrix in India. Approximately 560,000 dark factories on 1-acre coastal footprints are powered by 840 GW of dedicated zero-carbon nuclear baseload. Across a 7,500 km coastal corridor, these factories are projected to capture 28% of the world's physical manufacturing value-add. At 2024 world levels this is an after-tax profit of about $4.71 trillion per year, which is repaid to bondholders from the pooled net profit. Bharat starts after Iron Nexus is operative in Year 15: its profit begins in Year 19 and reaches full scale in Year 27 ($1.92T on the demand-gated case).
• Delper Aegis ($247.7 Billion ARR): An enterprise-tier industrial cybersecurity layer. Each production line sits behind an inline Delper Guard firewall targeting IEC 62443 Security Level 4 (SL-4), with a physical write-lock that blocks remote programming exploits.
• Delper Global Trust Infrastructure: A buyer-first verified sourcing platform. Track A (Bulk) carries a 15.184% fixed transaction rate, and Track B (Retail) carries 24.2%. Both run on a 14-day settlement cycle.

Deployment Curve

[Phase 1: Years 1-3]    ──> Digital AI Customs Gateways & Aegis 28nm/65nm Nodes Live (hardware); Aegis Revenue from Year 7; Track A & Track B Launch
                                ↳ Insulated by a 12-Month 100% Principal Exit Pipe (Section 5)

[Phase 2: Years 4-7]    ──> Coastal Automated Seaports Open; First 50,000 Factories Engaged

[Phase 3: Years 8-11]   ──> Electrified Rail DFCs & Inland Dry Ports Interlock; 300,000 Factories Active

[Phase 4: Years 12-15]  ──> Air Cargo Hubs Complete; 560,000 Factories / 840 GW Nuclear at 100% Scale

Human Capital & Cyber-Physical Defence

• 560,000 Forward Deployed Engineers (FDEs) per year: Each dark factory hub runs 7 FDEs for 24/7/365 monitoring across three 8-hour shifts. Training runs through a 12-month, online, credit-bearing Post-Graduate Advanced Diploma in Forward Deployed Engineering and Cyber-Physical Security, delivered via UVCE, Bangalore University, and scales to 2,147 physical hubs by Year 3.
• Coastal Defence Umbrella: The Indian Navy enforces a Maritime Domain Awareness bubble. Mid-tier lanes are covered by BrahMos batteries, and perimeters are protected by the Central Industrial Security Force (CISF) with anti-drone interceptor domes and air-gapped private 5G cores.

↑ Index

3. Integrated Capital Architecture: $102.00 Trillion

Total capital required is $102.00 Trillion , ring-fenced across eight asset layers over a 15-year construction horizon, funded at $6.40 Trillion per year .

The $102.00 Trillion splits into Project Iron Nexus ($46.00 Trillion) and Project Make in India 🇮🇳 ($56.00 Trillion).

Total Outflow: $102.00 Trillion over the 15-year build, equal to the sum of the eight asset layers above ($6.40 Trillion per year).

Investment and Pool Contribution by Stream

Repayment of both bond series is from the shared net profit pool, Iron Nexus first and then Bharat.

Funding the Total Outflow

The $102.00T outflow is funded entirely by bonds. Iron Nexus bonds carry a further $6.59T of capitalised construction interest, bringing their balance to $52.59T before repayments. bulk (Track 1) and retail (Track 2) repayments in Years 1 to 14 reduce it to $43.12T at Year 15 (Section 8). Both series are retired from pooled net profit.

↑ Index

4. Immediate Capital Ask & Strategic Use of Funds

The USD 8.00 Million investor cash is restated to match the costed build plan. Items marked "to be priced" are held until quotes are received, and the reserve is released only on Group approval. This restated split replaces the earlier Track A and Track B allocation, and it needs the investor's agreement before it is final.

Costed build, internal team: $0.987M

• Track A build [$0.485M]: architecture and control design, policy rules, data residency and consent, enforcement layer, observability, security baseline.
• Track B build [$0.502M]: agent build, channel connectors, kill switch and finance-lock tests, independent control review, soft launch in UAE and Singapore.

External items, to be priced: $2.915M

• SAP and systems [$0.915M]: SAP BTP and Signavio process mapping, SAP S/4HANA core ERP, and SGS quality-control webhooks, previously in Track A core systems and business process capital.
• 3PQC infrastructure [$0.80M]: partner onboarding, structural audits and independent quality assurance.
• Escrow, legal and licensing [$0.50M]: bank partner compliance, cross-border contracts and regulatory clearances.
• Supplier RTB vetting [$0.40M]: factory onboarding and vendor verification.
• Cloud and integration [$0.30M]: Shopify, WooCommerce and ONDC buyer endpoint connectors.

Operational contingency: $0.40M

Capital cushion against transaction latency in early processing loops.

Pre-shipment escrow margin: $2.60M (recycled, not spent)

A revolving liquidity margin on consecutive FOB payments. It is working capital, not a cost, and it is returned as payments recycle.

Reserve, held pending Group approval: $1.098M

Covers team capacity not yet costed in the build plan. It is released only on Group approval, once the build plan is extended or quotes are received.

Check: $0.987M + $2.915M + $0.40M + $2.60M + $1.098M = $8.00M.

Board approval resolution (template)

Resolution DG-BRD-2026-PH1-PROC, Delper Group (Top-Co), Dubai. Authorises the Phase 1 procurement suite (RFPs 1 to 12), the seed capital allocation, and Addendum A.

Recitals: Delper Group is executing Phase 1 of the 297-task Launch Master Plan ahead of the Month 7 Wave 1 soft launch in the UAE and Singapore. The Board has reviewed the restated USD 8.00M seed use of funds, the Master Procurement Calendar, the 100-point vendor scorecard and Addendum A. GK Bharta holds sole treasury, payment disbursement and unilateral kill-switch authority.

• Resolution 1, procurement suite: The 12-RFP calendar (including RFPs 3A and 3B) is approved for phased issue by the CTO, CBO, CCO and CFO, under HCLTech SIAM oversight and the Deloitte-provided CEO. RFP 13 (the Phase 2 global live channel, Section 14A) is not approved by this resolution and needs a separate Board resolution.
• Resolution 2, Addendum A and gates: No RFP is issued and no MSA is signed without Addendum A verbatim: the kill switch, GK Bharta Finance sole payment authority with a 48-hour release, in-country data residency, immutable audit logs with observer access, and 100% IP assignment. Bids are graded on the 100-point scorecard, and any bidder failing a pass/fail gate is disqualified.
• Resolution 3, seed tranche and reserve lock: The active operational tranche of USD 4.302M is released on milestone and quote verification: internal costed build USD 0.987M, external items USD 2.915M (SAP and systems USD 0.915M, 3PQC USD 0.800M, escrow, legal and licensing USD 0.500M, supplier RTB vetting USD 0.400M, cloud and integration USD 0.300M), and contingency USD 0.400M. The USD 2.600M escrow margin is recycled working capital and may not fund overhead. The USD 1.098M reserve is locked and can be drawn only by a further Board resolution, including any unbudgeted LMS (RFP 9) or QKD (RFP 10) commitment. Neither the seed nor the reserve can fund RFP 13 (the Phase 2 global live channel).
• Resolution 4, pre-launch validation: Wave 1 go-live requires written Board certification in Month 6 or 7 of (a) a live kill-switch and 48-hour payment-lock stress test across all 10 agents, (b) zero unresolved critical or high findings from the Trail of Bits and ConsenSys audits, and (c) CCO sign-off on VARA and MAS sandbox readiness.

Signatories: GK Bharta (Principal and Controlling Owner); the Deloitte-contracted CEO; an Independent Director; and the Chair of the Audit and Compliance Committee. Signature, name and date lines are left blank for execution.

↑ Index

5. Investor Risk Mitigation: The 12-Month Exit Pipe

The USD 8.00 Million investor commitment is insulated by a parallel commodity monetization facility that operates independently of the core technology tracks.

The isolated financial vehicle for GK Bharta

• Route: 1,000,000 metric tons of bulk palm oil annually, via irrevocable at-sight letters of credit (LCs).
• Revenue: GK Bharta earns USD 15.00 Million of post-tax palm oil revenue within the first 12 months.

Seed repayment and equity transfer

Palm oil revenue from GK Bharta funds the Year 1 retirement of the USD 8.00 Million investor loan, including 11% interest, totaling USD 8.88 Million, and the USD 4.00 Million Radiant Synage payment. The investor's 40% and Radiant Synage's 60% stakes are sold to GK Bharta at Month 12, so GK Bharta is the 100% owner from Year 2. Net cash risk to the investor drops to USD 0.00 at Month 12. Cash distributions to GK Bharta begin only after all loans are cleared, in Year 40 on the base case (Year 41 on the demand-gated case).

Seed cash flow (USD millions)

Track A and Track B allocations are itemized in Section 4. Group cash flows from Year 1 are in Section 9.

Post-launch buyout and equity transfer checklist (Months 8 to 12)

Confirmed: the Radiant Synage and investor buyout takes place in Month 12, at the end of Year 1.

Document DG-GOV-2026-M12-BUYOUT. Uses the isolated 1,000,000 MT a year bulk palm oil facility (irrevocable at-sight LCs). Target post-tax inflow USD 15.00M; coverage floor USD 12.88M. Controlling principal GK Bharta; independent witness EY.

↑ Index

6. Primary Unit Economics & Revenue Engines

Track A: Bulk Enterprise Trade

• Fixed Transaction Rate: Captures a mandatory 15.184% fee on total Gross Merchandise Value (GMV).
• High-Velocity Turnover: Achieves 26 capital turns per year via a 14-day settlement cycle from Year 2 onward (17.6 turns in Year 1).
• Liquidity Spread: Generates a 6.0% net annual spread on asset-secured capital facilities using bonded cargo as collateral.

Track B: Verified Digital Marketplace

• E-Commerce Take Rate: Captures 24.2% gross profit per order ($72.67 on a standard $300 customer checkout).
• Real-Time Bidding (RTB) Model: Buyers search free; pre-vetted factories prepay RTB and physical Third-Party Quality Control (3PQC) costs at actual cost, preventing platform overhead.
• Deducted Commissions: Platform commission and lead generation costs are deducted directly from vendor escrow balances only upon consumer satisfaction sign-off. Note that this is distinct from the 2% Transact Tax in Section 7, which is a protocol-level tax on token mint and burn.

Track 1 and Track 2 cash builds Aegis in Years 1 to 3. From Year 11, Track 1, Track 2 and Aegis revenue services and repays the loan as quickly as possible, and Iron Nexus income begins in Year 15. GK Bharta receives no profit share until all loans are cleared; from then on it receives 100% of net profit.

Delper Aegis: Factory-as-a-Service (FaaS)

• Platform ARR: Delper Aegis carries a base ARR of $247.7 Billion at full scale, deployed through 1,963,334 Delper Guard inline hardware units across SME and enterprise tiers.
• Tiered Hardware Margins: Targets 68% gross margin on SME chips and 82.3% gross margin on advanced 5nm dark factory chips.
• Insulated Staffing Markup: Pass-through engineer costs are absorbed by the factory customer, while Delper retains a 20% clean profit markup on all field operations.

Customer targets from Year 7

Aegis revenue starts in Year 7. The customer targets below are measured from that start, and the year references are operating years.

Bottom-up Aegis pricing and ARR model (full scale, Year 10)

This model builds the USD 247.70B Base ARR from customer segments and unit prices, since the blueprint notes the top-down figure has not yet been built from per-customer pricing. The bottom-up total is USD 247.78B, confirmed by management as the Year 10 full-scale figure.

Segment economics: Enterprise FaaS and hardware together give USD 74.98B, with USD 6.32B gross profit on the 5nm hardware at 82.3% margin. FDE markup gives USD 5.60B of operating profit. The SME tier gives USD 104.90B combined, with the 1,003,334 SME nodes at about USD 4,086 each (USD 4.10B, 68.0% margin). The home and student tier relies on B2B2C bundling with carriers, OEMs and Sana Learn accounts, which keeps acquisition costs near zero.

Ramp, Years 7 to 10

↑ Index

6A. Delper Global Retail (Track B) Financials

This section adds the financials of Delper Global Retail from the Global Verified-Commerce Trust Layer executive summary. Retail is Track B, funded by $4.00M of the investor's $8.00M. Its model ramps from 1.2 billion buyers in Year 1 to 6.0 billion in Year 5 . Figures are in USD billions unless stated.

21-year retail cash flow ($B)

Years 1 to 6: 60% of net profit is reinvested into retail working capital and 40% repays the loan. From Year 7, 100% of net profit repays the loan. Gross profit is flat from Year 5 and stays flat to Year 21. Year 1 is the model's ramp year and is not the India pilot, which is a separate 300 to 500 order test.

Unit economics per $300 order

Source: Global Verified-Commerce Trust Layer executive summary, sections "Rebuilt five-year forecast", "Loan repayment, Years 1 to 21" and "Group retail division: Track B financials". These are model outputs under stated assumptions, not forecasts, and have not been independently verified.

↑ Index

7. Macro Financing: Project Iron Nexus & Make in India 🇮🇳

The $102.00 Trillion programme is financed through the Delper Token economy, a supranational debt wrapper, and a fixed set of annual inflows:

The AAA Supranational Credit Enhancement Wrapper

• The Deflationary Mechanism: Standard emerging-market debt pricing at 9.00% would compound construction liabilities far beyond the blueprint's launch balance.
• Supranational Joint Guarantees: The World Bank Group (IBRD) and the Asian Infrastructure Investment Bank (AIIB) co-back an unconditional AAA guarantee over the construction tranches.
• Yield Compression: The model keeps the 9.00% rate on both bond series, so the wrapper compression to 4.75%–5.50% is not applied. Final launch debt at Year 15 is $43.12 Trillion , after Years 1 to 14 repayments from bulk (Track 1) and retail (Track 2), including capitalized construction interest, refinanced as a 10-year bond.

The Macro Toll Engine

Once fully operational in Year 15, the network collects an optimized $3.98 Trillion gross annual toll . After a $270.27 Billion annual O&M budget , this delivers $3.71 Trillion in net operational toll cash flow :

• Maritime Channels (35%): $1.39 Trillion per annum via automated TEU handling and deepwater berth access.
• Rail & Highway Tariffs (32%): $1.28 Trillion per annum ($0.15/km truck tariffs; $0.02/cargo-ton-km rail tariffs).
• Digital AI Customs Platforms (18%): $0.72 Trillion per annum generated from automated declaration processing fees.
• Inland Warehousing & Air Hubs (15%): $0.59 Trillion per annum via climate-controlled staging and smart sorting.

Annual Inflow Pipeline (Named Streams)

• Permanent Reserve Float: A $12.11 Trillion fiat reserve held in tier-1 banks at 4.00% returns $484.40 Billion per year toward debt retirement during the buyback period.
• 2% Transact Tax: A 1% tax on fiat deposit (buy) and 1% on fiat redemption (sell). On projected volume of $40.37 Trillion per year, this yields $807.40 Billion annually .

↑ Index

7C. Make in India: Updated Position (Demand-Gated)

This update applies the summary's stated Make in India profit schedule (Years 19, 23 and 27) with the demand-gated factory count (228,500 from Year 14, 40.8% of the 560,000 plan). The full ledger is in Section 9, and the stand-alone cash flow is in 7B.

On the demand-gated case, Make in India profit does not cover its own bond service. The shared pool, funded by Track 1, Track 2, the named streams and Aegis, pays the rest. The Make in India schedule in 7B shows the gap year by year.

Figures are model outputs under the summary's assumptions, with the demand gate from Section 7A. They are not independently verified. The profit basis (after tax or value-added share) and the stated-versus-ledger profit timing are both unresolved.

7A. Independent Review: Make in India Demand, Cash and Ramp

Added review. These findings are tests of the figures in Section 7 and 9. They use the summary's own numbers plus placeholder demand assumptions, and they have not been verified against external data.

What Section 9 shows for Make in India

• Bharat profit: $0.42T a year in Years 19 to 22, $2.52T in Years 23 to 26, and $4.71T from Year 27, all paid into the shared pool.
• Issue and repayment: $56.00T issued at the end of Year 29 and cleared in Year 40. Interest runs at 9% from Year 30, about $5.04T a year at the start.
• Distributions: to GK Bharta from Year 40, after both bond series are cleared.
• Integrity: the balance sheet check is zero in every year, and total inflow reconciles to debt service, tax, O&M, reserve, Aegis build and distributions.

Because Bharat is repaid from the pooled inflow, Make in India does not need to cover its own interest in isolation. The pooled model is the test that matters, and it clears by Year 40 on the summary's base assumptions and by Year 41 on the demand-gated case.

Demand-gated production ramp (world, placeholder assumptions)

Factory counts are the summary's India build plan. Demand is the consumer spend in eight markets with usable data (India, United States, Brazil, Mexico, Egypt, Pakistan, Nigeria, Bangladesh), with 15% of household spending in scope and adoption rising from 5% in Year 1 to 35% by Year 10. Each factory is assumed to produce $8.41M a year.

Demand is the binding limit from Year 10. Reaching 560,000 factories needs about $4.71T of demand a year, which at a 15% addressable share means roughly 86% of adults in these markets buying, against 35% assumed. The US accounts for about half of the Year 10 demand, and tariffs and domestic competition are not modelled.

Inconsistencies to resolve before reliance

• Profit basis: the $4.71T full-scale figure matches roughly 28% of global manufacturing value-added. If that is the basis, it is revenue-like rather than after-tax profit, and the Bharat inflows in Section 9 are overstated.
• Factory counts: the 80,000-factory global plan (960,000 lines) and the 560,000 India dark factories are not reconciled. The summary asks whether they overlap.
• Staffing: at 7 forward-deployed engineers per hub, 560,000 hubs would need about 3.9 million staff, while the summary also states 2,147 hubs by Year 3. The counting unit (factory, hub or line) must be fixed.
• Power: 840 GW of nuclear baseload is far above India's installed nuclear capacity today, so the power build is a critical path item.
• Buyers: 6.0 billion buyers equals about 72% of the world's population, and about 6.3 billion adults. Section 6 should use adult populations and country-level access instead.
• Years 1 to 14 funding: the build is funded by Iron Nexus flows and retained reserves. The $8.00M investor seed excludes Make in India, and no further round is provided.

Recommended gates for the Make in India ramp

1. Release each factory tranche only when confirmed orders cover at least 90% of its output, by product and destination country.
2. Restate the Bharat profit schedule on an after-tax basis, built from revenue, input costs and tax, before it is used in Section 9.
3. Confirm the power and transport build for each tranche before the factories in that tranche are commissioned.
4. Re-run Section 9 with the restated profit basis and a demand-gated ramp, and report the year Bharat is cleared under each case.

↑ Index

7B. Make in India Cash Flow, Years 1 to 48 (Demand-Gated Case)

Built from the demand-gated ledger in Section 9. Amounts in USD trillions. Stand-alone net = Make in India profit minus Bharat debt service. Years before issuance have no Bharat service. Pool cash from other streams funds any remaining shortfall.

↑ Index

8. Debt Retirement Schedule

At Year 15, construction tranches consolidate into two bond series, both carrying 9% per annum interest and issued as 10-year bonds. Bharat bonds are $56.00 Trillion. Iron Nexus bonds start at $52.59 Trillion ($46.00 Trillion principal plus $6.59 Trillion of capitalised construction interest). Repayments in Years 1 to 14 from bulk (Track 1) and retail (Track 2) reduce this to $43.12 Trillion.

Iron Nexus is bought back with 40% of Year 15 group inflow, and bulk (Track 1) and retail (Track 2) repayments reduce it from Year 15 on. From Year 16, the pool pays interest on Iron Nexus and then Iron Nexus principal. Bharat interest is capitalised into the Bharat balance while Iron Nexus is outstanding. Once Iron Nexus is cleared, Bharat interest and principal are paid from the pool. Nothing is distributed to GK Bharta until both series are retired.

Base Case: 100% Growth Track

Base case: Iron Nexus is retired in Year 29, and Bharat is issued at the end of Year 29 and cleared in Year 40. Demand-gated case: Iron Nexus is retired in Year 28, Bharat is issued at the end of Year 28 and cleared in Year 41. Repaid amounts include bulk (Track 1) and retail (Track 2) repayments.

Sensitivity: 130% Growth Track

Group inflows for Years 1 to 19 are scaled to 130% of the base case. Global Retail repayments are unchanged, and both series are repaid from the same pool. Years 20 onward are unchanged from the base case.

130% track, rebuilt on the financial statement basis: Iron Nexus is retired in Year 28 and Bharat in Year 38. Group inflows are scaled to 130% in Years 1 to 19; distributions to GK Bharta total 116.50T through Year 51.

Post-Debt Capital Allocation

Once both bond series are retired, 100% of net profit is distributed to GK Bharta. Distributions to GK Bharta start in Year 40 on the base case, and Year 41 on the demand-gated case, after Bharat is cleared. On the 130% track, Iron Nexus is retired in Year 28 and Bharat in Year 38.

↑ Index

Demand-gated case (Years 15 to 46)

Same sequence as the base case, with Make in India profit on the summary's stated schedule (Years 19, 23 and 27), limited to the demand-supported factory count from Year 14. Iron Nexus is retired in Year 28 and Bharat is cleared in Year 41. The full ledger is in Section 9.

9. The 51-Year Consolidated Cash Allocation Ledger (Demand-Gated Case)

This ledger replaces the summary's Section 9 ledger. It applies two changes: Make in India profit follows the summary's stated schedule (Years 19, 23 and 27) and is limited to the demand-supported factory count from Year 14 (228,500 factories, 40.8% of the 560,000 full-scale plan), and the model is rebuilt so every year balances. Other streams (Track 1, Track 2, named streams, Aegis and O&M) are unchanged. The summary ledger is kept for comparison below.

Base case against demand-gated case

The demand-gated case keeps the $56.00T Bharat bond unchanged. A bond sized to a smaller build would reduce this amount, but the summary does not provide a resized bond, so the principal is held as stated.

Year-by-year ledger (demand-gated case)

Basis and assumptions

• Make in India profit: the summary's stated profit schedule ($0.42T from Year 19, $2.52T from Year 23 and $4.71T from Year 27) is applied as written, then scaled by 40.8% from Year 14 to reflect the demand-supported factory count of 228,500. The ledger's own timing (Years 25, 29 and 33) differs from the text and is not used in this case.
• Iron Nexus: $52.59T balance at Year 1 (principal $46.00T plus $6.59T capitalised interest), repaid from the pool. Interest at 9% on the opening balance from Year 16. Retired in Year 28 under the demand-gated case.
• Bharat: $56.00T issued at the end of Year 28, once Iron Nexus is cleared. Any leftover pool cash in that year reduces the Bharat balance. Interest at 9% from the following year. Retired in Year 41.
• Tax: 17% of EBITDA less interest, paid from the pool before debt repayment. Depreciation is not charged.
• Retained reserve of $0.03T a year in Years 1 to 10. O&M of $0.27T a year from Year 16.
• Distributions to GK Bharta start only after both bond series are cleared, in Year 41.
• Model check: total inflow equals O&M, tax, debt repaid, retained reserve, Aegis build and distributions in every year. The balance sheet check is zero in every year.

The demand-gated case assumes factories are built to the demand-supported count. Build cost is not reduced, so the Bharat bond is unchanged. Financing of the Years 1 to 14 build is taken from the summary's Iron Nexus flows, which are not independently verified.

9A. Pooled Cash Flow (Demand-Gated Case)

Pooled cash flow for Years 1 to 51, by revenue stream. Total inflow equals O&M, tax, debt repaid, retained reserve and distribution in every year.

9B. Financial Statements (Demand-Gated Case)

Profit and loss

Balance sheet

Check = total assets minus bonds minus equity, zero in every year.

Debt service coverage

Vital financial ratios

↑ Index

9C. Enterprise Valuation at 20x Net Income (Demand-Gated Case)

This section applies a 20x multiple to each year's net income from Section 9B, giving an enterprise valuation for each year and the year-on-year change. The multiple is an assumption for this calculation, not a figure from the source documents. Net income is after interest and tax. The Section 9B figures are used as stated, and Bharat profit carries the basis issue noted in Section 7A.

Years 1 to 41 match the Section 9B horizon. Values are in USD billions (B) and trillions (T), and the change column is the difference from the prior year.

↑ Index

10. Strategic Investment Summary, Validation Gates & Milestones

• Blueprint Verdict: The Delper Integrated Master Blueprint states the programme is unconditionally viable, citing the token economy's 2% transaction tax, the $12.11 Trillion treasury float, and 24/7 software-defined manufacturing as drivers of rapid capital amortization.
• Post-Debt Treasury: Once debt is retired, the project operates at macro profit margins, leaving the corporate entity an unencumbered treasury to capitalize future global trade expansion.
• Immediate Milestone Gating: The multi-trillion-dollar expansion is insulated behind short-term operational proof gates:

[Month 0-2: Framework Validation]  ──> Secure letters of intent from Axis Bank, ONDC, and SGS.

[Month 2-9: Cost-Per-Order Pilot]  ──> Run 300-500 transactions to verify supplier fee acceptance.

[Year 2-5: Network Scalability]    ──> Scale to 15,000 sellers before advancing Aegis FaaS capital.

Non-Negotiable Validation Gates (First 36 Months)

1. Industrial Frame Stress Testing: Inject over 5,000,000 malformed frames per hour into 5nm FinFET Delper Guard processors to confirm self-quarantine without dropping robotic synchronization.
2. 60-Day Shadow Mode: Passively record the first 4,000 Year 1 factory hulls for at least 60 days to build baseline models of Modbus/PROFINET traffic before live enforcement.
3. Wafer Procurement Lock-In: Secure multi-year advanced node fabrication agreements with global foundries against semiconductor supply shocks.
4. Legal Restructuring of Training Bonds: Complete a multi-jurisdictional review of student repayment terms so that ₹20.0 crore per-hub refund exposure is legally isolated from tuition revenue.
5. Group Software Human Override Policy: All Delper software that makes or executes decisions affecting customers, traders, workers, public infrastructure, or safety must have a designed and tested human override before go-live. Each override needs a named owner, defined thresholds for when it applies, and an audit log of every use. Agentic systems, including customs clearance, must define in writing the actions they may take without approval and the actions reserved for a named human, such as customs holds, rejections, duty assessments, enforcement, and safety-related operations. An emergency halt must be tested, with a documented manual fallback. The 60-day shadow mode must measure override frequency and response time as well as detection accuracy.

Five non-negotiable validation gates (Months 1 to 36)

Year 7 Aegis economics and workforce targets

• Margins: USD 247.7B Base ARR at full scale; 68.0% gross margin on standard SME chips and 82.3% on 5nm dark-factory chips; 20.0% clean markup on pass-through field engineer salaries.
• Security: Inline Delper Guard firewall targeting IEC 62443 SL-4, with a physical write-lock and a sealed quarantine store driven by 52 multi-domain threat baselines.
• FDEs: 560,000 trained a year through a 12-month post-graduate diploma, across 2,147 hubs by Year 3, staffing dark factories 24/7/365.
• Customers (Years 7 to 15): 80,000 FaaS factories (960,000 lines) by Year 10, 70% of global SMEs and 6.0 billion home and student users by Year 12.

↑ Index

11. ESG and SDG Alignment

17 SDG global matrix summary

The matrix lists all 17 UN Sustainable Development Goals in ascending order, with each goal's contribution in this summary, where it is covered, and the evidence needed before the claim can be relied on. Goals marked "new to this matrix" were not previously mapped in this section. Contributions are design claims until they have a baseline, a metric and an independent verifier.

This section maps the programme to the UN Sustainable Development Goals (SDGs) and to the Environmental, Social and Governance (ESG) pillars. It records where the programme contributes, where it creates risk, and what must be evidenced before alignment can be claimed. The mapping uses the figures stated in this summary and has not been independently assessed. Each contribution is a design claim until it has a baseline, a metric and an independent verifier.

SDG contribution map

An audit of the newly incorporated Global Ecosystem Regeneration & Utility Initiative (20-Year Master Plan) confirms that the introduction of the macro-CSR framework fully addresses and bridges the 7 previously excluded UN Sustainable Development Goals (SDGs). By funding this blueprint using a strict 2% corporate pretax profit engine, the Delper Group expands its footprint from isolated industrial hubs into a planet-wide utility network targeting 5.5 million village modules. The sections below show how the 7 previously missing SDGs are explicitly integrated and operationalised within the updated framework.

SDG 1: No Poverty

• Active Mechanism: The initiative directly introduces 11 million marginalized women into the grassroots economy as equity-holding micro-entrepreneurs running the 31,583 walk-in smart pods.
• Economic Impact: Injects $23.14 Million USD (Rs. 2,221 Million) in net monthly earnings directly into low-income communities.

SDG 2: Zero Hunger

• Active Mechanism: Pods provide instant, balanced meals 24/7 via specialized culinary partnerships (Shastry's Kitchen and Shetty's Kitchen).
• Macro Scale: Connects 5.5 million solar cold-storage nodes to an AI-driven predictive logistics marketplace. This eliminates predatory middleman markups, aims to recover $1.2 Trillion USD in annual post-harvest crop spoilage, and directly targets the structural food deficit.

SDG 3: Good Health and Well-being

• Active Mechanism: Every rural phase module deploys a telehealth diagnostic room that connects remote villagers to urban specialists.
• Execution: Integrates over-the-counter wellness shelves inside smart pods and establishes automated telemetry networks to maintain health utility uptimes.

SDG 5: Gender Equality

• Active Mechanism: The retail micro-commerce engine explicitly mandates a 100% female operator framework for the global urban pod rollout.
• Execution: Shifts local financial agency to marginalized women by giving them direct ownership stakes in localized digital banking and biometric micro-ATM nodes.

SDG 6: Clean Water and Sanitation

• Active Mechanism: Deploys 5-stage RO/UV water ATMs hooked up to 200-litre standalone battery-backed tanks.
• Macro Scale: Funds deep desilting of community ponds, constructs rainwater harvesting matrices, implements greywater natural reed-bed filtration channels, and deploys water-free composting bio-toilets across 5.5 million villages.

SDG 10: Reduced Inequalities

• Active Mechanism: Replaces volatile, short-term charity models with a zero-cost infrastructure deployment model for beneficiaries.
• Execution: Grassroots communities assume zero debt and zero operating obligations. The 2% corporate profit engine distributes asset ownership downward, lowering productivity losses and baseline structural wealth gaps.

SDG 11: Sustainable Cities and Communities

• Active Mechanism: Stage 1 installs dense, weather-sealed 8 ft by 6 ft by 8.5 ft smart pods inside high-density, low-income urban settlements and transit points.
• Macro Scale: Ramps into Stage 2 (Years 10–20) to retrofit village arrays into continental smart energy grids, connecting isolated rural zones to national power corridors via bi-directional edge inverters.

Updated 17 SDG global matrix summary

ESG pillar review

Actions before ESG and SDG claims are relied on

1. Impact assessment: Commission an independent environmental and social impact assessment for the coastal corridor, nuclear sites and rail routes.
2. Carbon baseline: Publish a carbon baseline and a net-zero construction pathway that covers cement and steel.
3. Palm oil policy: Adopt a sourcing policy with certification, traceability and deforestation-free commitments before the palm oil route begins.
4. Labour and community framework: Set a labour, human-rights and community-engagement framework covering workers across the 560,000 dark factories and affected communities.
5. Financing evidence: Obtain written confirmation of the IBRD and AIIB guarantee terms, and an independent legal review of the token-based financing.
6. KPIs and reporting: Define 8 to 12 indicators for the aligned SDGs, each with a baseline, a target and a verifier, and report annually against GRI or ISSB (IFRS S1 and S2).
7. Group software override policy: Adopt a Group-wide policy that every system making or executing decisions affecting customers, traders, workers, public infrastructure or safety has a designed, tested human override, with a named owner, defined thresholds, and an audit log of each use. Agentic systems also need written limits on their permitted actions, appeal routes for traders, change control for model updates, and a named executive accountable for their decisions. Cover this in the SDG 16 and governance review.

Alignment statements in this section are design-level assessments based on the figures in this summary. They are not an external ESG rating or SDG certification, and they have not been independently verified.

↑ Index

12. Activities to Achieve ESG and SDG Alignment

The activities below are the work required before the ESG and SDG claims in Section 11 can be relied on. Each one is tied to the SDGs it supports and to the evidence it will produce. Owners and dates are to be confirmed by the Group.

Environmental

Social

Governance

Measurement and reporting

↑ Index

13. Learning Platform: Sana Learn

Sana Learn is proposed as the learning management layer for onboarding, training and compliance across the Delper Group. It combines an LMS and a learning experience platform in one AI-native system. Its role is to deliver and track training. It does not replace the university, engineering or operating systems described elsewhere in this summary.

Features relevant to Delper

• Course authoring and assessments: course builder, quizzes and tests, and support for instructor-led sessions.
• Content conversion and import: PDFs can be converted into interactive courses, and existing content can be imported via SCORM.
• Automated enrollment: programs with enrollment rules, reminders, and certificate generation.
• HR system integration: automatic enrollment and in-person session scheduling linked to HR records.
• Virtual classroom: polls, quizzes, reflection threads and breakout rooms for live cohorts.
• AI tutoring and dashboards: personalized learning paths and performance reporting.
• Mobile and microlearning: short modules suited to field and factory staff.

Where it fits

Cost and scale

Pricing is not confirmed. Published figures conflict, ranging from $13 per user per year to $13 per user per month, with some listings starting at $8. At $13 per user per year, 1,000 learners cost about $13,000 a year. The 560,000 FDEs would cost about $7.3 million a year, or about $87 million a year at the monthly rate. The Track A and Track B use-of-funds tables have no LMS line, so this cost must be budgeted or negotiated separately.

Leadership training for managers

Managers who oversee AI-driven onboarding and sales need training in leading people through change and in setting AI governance controls. We recommend two MIT Sloan Executive Education courses, with enrollment subject to MIT's own admission process:

• Managing the Human Side of Digital Disruption (primary): A six-week, self-paced online course with live MIT faculty sessions, covering how roles, capabilities and emotions change during digital transition. It can be bought for a whole team.
• Implementing Agentic AI: Building Your Organizational Playbook (secondary): A short online course, with no technical background needed, that produces a governance framework and a pilot plan. It supports the design of the parameters, escalation rules and human override controls described in Section 14.
• Designing Organizations for Agentic AI (Harvard Business School Online, complementary): A four-module online course of 20 to 25 hours, with 90 days of access. It covers evaluating and deploying AI agents, redesigning workflows for human-agent collaboration, and applying governance, security and AI sovereignty frameworks. It suits the executives who will oversee the Deloitte-provided CEO function and the agent governance controls in Section 16.

Course details come from MIT Sloan Executive Education and GetSmarter. Training supports the governance controls in this section but does not replace them.

Sana feature and pricing information comes from Sana's published materials and third-party reviews. It has not been verified in a product demonstration, and pricing should be confirmed with the vendor.

↑ Index

14. Global Onboarding

Partners, operators, suppliers and team members joining the Delper Group across regions can start their onboarding once the Group confirms the sign-up route for their market. Onboarding steps are confirmed once the sign-up is complete.

Onboarding terms, eligibility and any regional requirements are set by the Group and the applicable jurisdiction. This link does not constitute an offer of securities or a financing commitment.

Parameter-based automation with human access

Delper's onboarding and sales processes run on defined parameters. Each step follows written rules and thresholds, such as eligibility criteria, document requirements, consent status and approved outreach templates. The system acts only within those parameters. Any case that falls outside them, or that the rules cannot decide, is routed to a person. Every automated process that makes or executes a decision affecting a partner, supplier, operator, team member or prospect has a named human owner, as set out in the Human Override Policy in Section 12.

At any point in onboarding or sales, an applicant, prospect or partner can choose to connect with a human. The request is routed to a named team member, and the person can pause or change the automated steps. Sign-up creates the partner or supplier record in Delper's own systems, and the sign-up route is to be confirmed by the Group.

In sales, AI searches Apollo for prospects that match a Delper service, within Apollo's terms and rate limits. The workflow drafts outreach only for prospects with a lawful basis for contact in their market. A salesperson approves each message before it is sent. Each message identifies Delper and includes an opt-out, and opt-outs are kept in Delper's own CRM and applied across all tools.

Procurement suite: RFPs 1 to 12

RFP 5 and RFP 7 detail

RFP 5, smart contract audits and penetration testing. Ahead of the Month 7 soft launch, Delper seeks dual independent audits (manual review, formal verification and symbolic execution) covering the 1.998 trillion token cap, USD 21.40 price floor, 0.5% buy and 0.5% sell tax, the 6-layer recursive contracts (4.15× multiplier), the ETH stability pool and the kill-switch privileges. It also requires full-stack penetration testing of mobile apps, web portals, APIs and SAP connectors, including 5,000 TPS load and DDoS resilience, a public bug bounty on Immunefi or HackerOne, and remediation certificates for VARA, MAS, FCA and MiCA.

RFP 7, AML/KYC, KYB and on-chain fraud screening. Covers consumer KYC and biometrics (Aadhaar, PAN, Emirates ID, Singpass, passports, sanctions and PEP screening) inside the Layer 1 poll-to-wallet flow in under 15 seconds; corporate KYB with UBO extraction across six supply-chain tiers; on-chain KYT and FATF Travel Rule for DPR and ETH pool addresses; and human-in-the-loop routing of all holds and rejections to compliance analysts under the CCO, with in-country residency and GK Bharta suspension hooks.

Master procurement calendar (Months 1 to 12)

Immediate wave (Months 1 to 3): RFP 11 (legal and licensing counsel) and RFP 12 (PwC and EY) issue in M1 Week 1 and are awarded in M1 Week 4. RFP 2 (HCLTech SIAM) is awarded in M2 Week 1. RFP 4 (SAP core ERP) is awarded in M2 Week 3. RFP 6 (SGS 3PQC) is awarded in M2 Week 4.

Build and enforcement wave (Months 2 to 5): RFP 3A (control plane) is awarded in M3 Week 1. RFP 7 (AML/KYC) in M3 Week 2. RFP 5 (audits and pen test) in M3 Week 3. RFP 1 (10-agent platform) in M3 Week 4. RFP 3B (ElevenLabs guardrails) in M4 Week 1.

Commercial and future-readiness wave (Months 4 to 12): RFP 8 (payments and POS) is awarded in M5 Week 2. RFP 9 (LMS pilot) in M5 Week 4. RFP 10 (QKD hardware) in M11 Week 3, for a feasibility study in Months 12 to 14.

Critical path gates before Wave 1: Month 6 Week 2, kill-switch and finance-lock drill (USD 40,000 validation task). Month 6 Week 4, audit and pen-test clearance by Trail of Bits, ConsenSys and EY. Month 7 Week 1, CCO sign-off on UAE (VARA) and Singapore (MAS) sandboxes for Agents 1 to 9. Month 8 Week 4, Wave 1 pilot and buyout review, validating 50,000 pilot transactions.

100-point vendor evaluation scorecard

Any proposal that fails one of five pass/fail sovereign gates is disqualified before scoring: (1) Addendum A signed without redlines to core control clauses; (2) a documented GK Bharta kill-switch hook; (3) written acceptance of GK Bharta Finance sole payment authority and the 48-hour release schedule; (4) architectural proof of in-country residency for PII, financial records and audit logs (UAE, Singapore, India under DPDPA, UK and EU under GDPR); (5) written confirmation of zero independence conflicts with Deloitte, PwC and EY.

Award thresholds: 85 to 100, Tier 1, advance to MSA and Board disbursement approval. 70 to 84, Tier 2, issue a best and final offer. 55 to 69, Tier 3, reserve only. Below 55, disqualified and logged.

Year 2 global expansion and licensing tracker (Waves 2 to 4)

Document DG-REG-2026-Y2-LICENSING. Months 9 to 24 and beyond. Compliance lead: CCO and CLO. External partners: Cooley, WilmerHale and Clifford Chance (RFP 11), and PwC compliance management (RFP 12). No agent, payment rail or DPR liquidity pool goes live in a jurisdiction until its gate is signed off in writing by the CCO and recorded in Agent 8.

Market activation prerequisites: local data residency pods, automated PII and language redaction, licensed payment partners, restricted exchange trading pairs, and CCO approval, all signed off by Agent 8 and PwC. Year 2 budget: USD 1.50M to 2.50M for legal retainers (RFP 11) and USD 2.50M to 4.00M for compliance operations and audits (RFP 12), backed by a minimum USD 50.00M cross-exchange liquidity pool and monthly public ETH stability attestations by EY.

↑ Index

14A. Global Live Channel and RFP 13 (Phase 2 proposal, not in base case)

This section sets out a 24/7 mobile news and sports channel, delivered free to users in 195 countries and funded by sponsored mobile data and ad revenue. It is a proposal. It is not part of the Phase 1 plan, the seed, or the base-case ledgers in Sections 8 and 9. Adopting it needs a separate Board resolution, a funding source outside the seed, and a launch after the Month 12 buyout.

Channel costs (annual unless stated)

Recurring operating cost excluding rights is $72M to $145M per year.

Free data under a 500 MB daily cap

At 500 MB per user per day, annual traffic across 6.0 billion users is 1,095 billion GB (1,095 exabytes). Three hours of programming fit inside 500 MB only at about 370 kbps (240p or 360p, using AV1 or HEVC, roughly 166 MB per hour). At standard 480p, the cap gives about one hour of free viewing per day.

Ad revenue and net position

Three hours of ad-supported programming gives 36 to 45 ad impressions per user per day, or 78.84 trillion to 98.55 trillion impressions a year across 6.0 billion users.

The table below shows ad revenue less data cost, before the broadcaster split and operating costs.

At a $4.00 eCPM, break-even is 18.75 impressions per user per day at $0.15 per GB, and 37.5 at $0.30 per GB.

Assumption to confirm. The broadcaster split is 30% to 50% of the margin after data cost. This section uses 50% as the conservative case. At $4.00 eCPM and $0.15 per GB, Delper keeps about $75.4B to $114.9B a year after the split and operating costs. At $0.30 per GB, the margin is negative at the low end, so the channel is self-funding only at the high end (about $32.8B to Delper). At $2.00 eCPM, the channel is not self-funding at $0.30 or $0.60 per GB. Any surplus flows to the Delper pool, which repays Iron Nexus first (Year 29 base case; Year 28 demand-gated) and then Bharat (Year 40 base case; Year 41 demand-gated).

Funding

The seed cannot fund RFP 13. The USD 8M to 15M setup cannot come from the active tranche of USD 4.302M, the USD 2.600M escrow margin, or the locked USD 1.098M reserve. The only candidate source is the retained reserve of USD 30B a year in Years 1 to 10 (Section 9). Pool cash starts in Year 1, its use needs Group approval, and so it cannot fund setup in Months 4 to 8. The recommendation is to defer the RFP 13 award until Year 1 pool cash exists, and to keep the channel out of the Wave 1 soft launch in UAE and Singapore (PH 21, Month 7).

Scope notes

• Ad partners. The PH 17 partners cover merchant and consumer ad credits in Months 3 to 9. Broadcast ad insertion is new scope and needs contract amendments with each partner.
• Voice and dubbing. RFP 3B (ElevenLabs, USD 210,000 to 420,000) covers voice agents only. Live news dubbing is new scope and needs separate pricing.
• Money controls. Agent 10 is read-only. It reconciles impressions, MNO charges and the broadcaster split. GK Bharta Finance calculates and pays.

RFP 13 gates and scorecard

Bidders must pass five gates before scoring, the same gates as Section 10: (1) Addendum A signed without redlines to core control clauses; (2) a documented GK Bharta kill-switch hook that halts any feed in under 500 ms; (3) written acceptance of GK Bharta Finance sole payment authority and the 48-hour release; (4) architectural proof that viewer data, ad telemetry and MNO logs stay in-country under India DPDPA, Singapore PDPA, UAE regulations, UK GDPR and EU GDPR; (5) written confirmation of zero independence conflicts with Deloitte, PwC and EY.

The 100-point scorecard covers: ingest, codec and scale (30); SSAI and MNO enforcement engine (25); security, audit and privacy (20); commercial terms and TCO (15); delivery and IP assignment (10). Award thresholds: 85 to 100 is Tier 1 (advance to MSA and Board approval); 70 to 84 is Tier 2 (best and final offer); 55 to 69 is Tier 3 (reserve); below 55 is disqualified.

Proposed pre-launch tests

These run only if RFP 13 is funded and awarded before Month 6. Otherwise they move to the Year 1 cycle.

• TC-13A (bandwidth cap): free data cuts off, or shifts to audio, at exactly 500.0 MB per device per day, with no bill shock to Delper or the user.
• TC-13B (feed severance): GK Bharta can sever any broadcaster feed or ad exchange in under 500 ms, within the DG-QA-2026-M6-KSFL drill.
• TC-13C (reconciliation): Agent 10 reconciles impressions, MNO charges and the broadcaster split into SAP S/4HANA with zero discrepancy, before GK Bharta Finance releases payment under the 48-hour lock.

Board actions before any commitment

1. A new resolution adding RFP 13 to the approved calendar, since Resolution 1 covers RFPs 1 to 12 only.
2. Approval of the funding source for setup, from the retained reserve, with the timing in the Funding section above.
3. A decision on whether the TC-13 tests join Resolution 4 certification.

Open items

• RFP 6 reconciled. RFP 6 (SGS third-party quality control and supplier RTB vetting) is USD 1.2M. This equals the seed's 3PQC line of USD 0.80M plus its supplier RTB vetting line of USD 0.40M in Section 4. No change is needed.
• Channel assumptions. The ad, data and broadcaster-split assumptions above are unverified and are not part of the base case.

↑ Index

15. AI Ethics and Governance

Delper uses AI in onboarding, sales prospecting and operations. This section sets out the ethical commitments that apply to those uses. They are proposed commitments, to be approved by the Group before they are relied on.

Principles

• Human accountability: Every automated process has a named human owner, as set out in Section 12 and the Human Override Policy.
• Human access: Any applicant, partner, supplier, prospect or team member can ask for a person at any stage, and can request a human review of an automated decision that affects them.
• Explainable decisions: Each decision to accept, reject, suspend or delay an applicant records the parameters it applied, so the reason can be given in plain language.
• Consent and lawful basis: Outreach is sent only where a lawful basis exists in the prospect's market, under India's DPDP Act, GDPR and PECR. Each message identifies Delper and offers an opt-out, which is honoured promptly.
• Data minimisation: Only the data needed for onboarding, compliance and sales is collected, and it is held in Delper's own systems.
• Fairness: Eligibility and scoring parameters are tested for disparate impact across regions, genders and business sizes before use, and retested at least annually.
• Transparency about AI: Prospects and applicants are told when they are dealing with an automated process, and when a message was drafted by AI and approved by a person.
• Safety and security: Automated systems that touch factory operations use the hardware-enforced controls described in the Aegis summary, and are subject to a halt control.

Risks and controls

Governance

An AI ethics review is held before any new automated process goes live, and each automated process is reviewed at least annually. The review records the parameters, owners, escalation rules and fairness results. Reports go to the Group board. Leadership training in AI governance, set out in Section 13, supports these reviews.

Addendum A: Master Vendor Governance and Sovereign Control Rider

This rider is part of every RFP, SOW and MSA with Delper Group and its subsidiaries (RFPs 1 to 12). Where a vendor's standard terms conflict with it, the rider governs.

1. Unilateral master kill switch. GK Bharta alone may halt, pause, suspend or terminate any automated decision, agent workflow, API, data pipeline or vendor service, inside or outside parameters, on a documented trigger. Vendors expose an authenticated, low-latency halt hook and cease operations without penalty or SLA claim. For EY, GK Bharta may suspend or replace the engagement but may not alter its findings.
2. Sole financial control and 48-hour release. GK Bharta Finance alone operates banking, treasury, escrow releases and disbursements. No vendor, Deloitte or HCLTech SIAM holds payment authority. Every approved invoice has a 48-hour lead time from approval to release, and vendors may not throttle or suspend during that window.
3. In-country data residency. Customer, merchant, supplier, employee, telemetry and transaction records stay in their jurisdiction under India DPDPA, Singapore PDPA, UAE regulations, and UK and EU GDPR. Cross-border transfers, external model training on Delper data and vendor multi-tenant storage are prohibited unless the CCO and the GK Bharta Board authorise them in writing.
4. Immutable audit logging and observer access. Every automated action, policy check, escalation and configuration change produces a tamper-evident, exportable record. GK Bharta (read-only observer), HCLTech SIAM, PwC and EY receive real-time read-only access. Automated decisions record the parameter rules applied, to support plain-language explanation on review.
5. Human-in-the-loop and named accountability. Out-of-parameter, high-risk or disputed actions route to a named Delper owner, never an unmonitored queue, with a tested manual fallback. Vendor agents have no financial write or posting access unless the GK Bharta Board authorises it in writing.
6. IP assignment and transition. All custom software, orchestration code, smart contracts, webhooks, process maps, UI assets and fine-tuned model weights vest in Delper Group as works made for hire on milestone payment. On exit, vendors deliver documentation, revoke credentials and transfer data and source code within 30 days.

Vendor acknowledgment: the signing representative confirms acceptance of Addendum A without modification, with blank fields for vendor name, RFP number, signatory, signature and date.

↑ Index

16. Launch Master Plan: 297 Tasks

This section sets out the Delper Ecom Launch Master Plan, from incorporation to global launch. It contains 297 tasks across 22 phases. Timelines are in months from launch (M1 to M24 and beyond). The plan is proposed and has not yet been approved by the Group. Where its figures differ from the rest of this summary, the conflicts are listed below rather than changed silently.

Phase summary

Compliance: achieved and managed by PwC

Compliance will be achieved and managed by PwC. Financial audit is performed by EY. Delper pays both firms' fees. In Years 1 to 10 the fees are met from the retained reserve of $30B a year, and from Year 11 from pool cash before debt repayment. The compliance cost estimates in this section are placeholders until PwC provides a quote. GK Bharta does not pay compliance. The Iron Nexus clearance (Year 29 base case; Year 28 demand-gated) and the Bharat clearance (Year 40 base case; Year 41 demand-gated) are shown in Section 9. The use of the retained reserve for compliance requires Group approval.

Service integration: SIAM by HCLTech

HCLTech will act as Service Integration and Management (SIAM) provider. It coordinates the multi-supplier ecosystem, including PwC for compliance, EY for financial audit, the technology, payment and data vendors, and the partner network. It owns the service-level agreements, incident and dispute processes, and the reporting that links them to Delper's governance. HCLTech's SIAM service is charged as a separate fee, which is not included in the compliance or cost estimates in this section. HCLTech holds Scopism SIAM Assured accreditation, which it was first to achieve and to renew.

Tasks that fall within SIAM scope include the partner SLA monitoring and offboarding process (PH 19), the layer-by-layer settlement checks (PH 11 to PH 15), and the operational monitoring dashboards. The task list is unchanged.

Responsibility matrix (RACI)

R: responsible for doing the work. A: accountable for the outcome and the decision. C: consulted. I: informed.

Executive and reporting line: GK Bharta owns the Group. Deloitte provides CEO services under a contract with GK Bharta, and the CEO reports to the GK Bharta board. The board is the ultimate authority over the CEO and the business process. The CEO is accountable for day-to-day decisions, and the board can override them. Automated processes keep a named owner for overrides and halts, as set out in Section 12.

Independence: Deloitte is not the compliance provider or the financial auditor. PwC runs compliance and EY performs the audit. Deloitte should confirm that providing CEO services creates no conflict with those roles.

Observability for business process operations

Splunk's 2026 State of Observability report (2,050 ITOps and engineering professionals) sets out findings that apply to Delper's operations. The table maps each finding to a Delper implication and control. Observability tooling is not in the 297-task list, so adopting a platform would be a new task for the Group to approve.

Observer: GK Bharta acts as the observer of Delper's observability. It has read-only access to the telemetry, operations and cost dashboards, and receives the AI, security and SLA alerts. The observer role is oversight only: it doesn't operate the systems, and it doesn't replace the board's override rights over Deloitte-provided management. Delper's operating teams and HCLTech SIAM keep responsibility for running the monitoring.

Figures are from Splunk's State of Observability 2026 report and press coverage, and have not been independently verified.

Finance and payments

GK Bharta alone operates finance and makes all payments. No other provider in this plan, including Deloitte, HCLTech SIAM, PwC and EY, runs finance, treasury or payments, and none holds payment authority. Their scope excludes finance. EY performs the independent financial audit and does not operate the finance function.

Every outgoing payment has a lead time of 2 days from approval to release. This applies to all payments, including licence fees, audit fees, provider fees, salaries and distributions.

The finance tasks in PH 03 are carried out by GK Bharta. The task list is unchanged.

Kill switch

GK Bharta alone holds the kill switch. It can halt any decision made within or outside the parameters, and any activity of Deloitte, HCLTech SIAM, PwC, EY and the business process. Each provider's contract must give GK Bharta this right, and each provider must comply with it. The kill switch is exercised through a documented trigger and logged in the audit trail.

Cryptography and QKD pilot

Post-quantum cryptography is the baseline for the whole business process. It covers the ledger, APIs, partner links, the Aegis firmware signing and the data stores, and runs on existing networks and devices. NIST published its first post-quantum standards in 2024 (FIPS 203 for key exchange, FIPS 204 and 205 for signatures).

QKD is not applied across the whole process. It is limited to a pilot between the four hubs (Dubai, Singapore, Bengaluru/Mumbai and London), to be started only after the platform is live and proven. Scenario A of the pilot is estimated at USD 1.0M to 4.4M one-off and USD 0.1M to 0.3M a year in maintenance, before any tariff impact. Extension to a critical network of about 50 links (USD 16M to 90M) is subject to the pilot results and Group approval. Whole-process QKD, covering about 80,000 factory links, is not feasible at an estimated USD 16 billion or more in hardware.

These three tasks are proposals for Group approval and are not in the 297-task list. Costs follow the Scenario A estimate above. Payments follow the 2-day payment lead time, with GK Bharta finance making all payments.

Security assurance statement

The Delper global business process is designed to be resilient and recoverable. It is not represented as hacker-free. QKD protects key exchange on the links it runs over, and detects eavesdropping on those links. It does not protect endpoints, software, trusted relay nodes, availability or people. Security therefore rests on the layered controls below, not on QKD alone.

Self-quarantine: Automated quarantine is the containment layer. It acts within seconds to isolate a threat, and the GK Bharta kill switch remains the final halt. The Aegis summary sets the 3-second containment target and cites CrowdStrike figures, including an average eCrime breakout of about 29 minutes. These figures are as quoted in the Aegis summary and need confirming against primary sources. The Aegis summary also excludes its self-quarantine segment from several of its own financial sections, so it isn't used to size the investment here.

The aim is detection, containment and recovery without loss of money or control. No control set can guarantee that no attack succeeds.

Agentic AI control plane (SentraMesh)

SentraMesh describes itself as an AI governance and control platform for autonomous agents. It sits between agents and the systems they call, providing visibility, inline policy enforcement, behavioural risk scoring and an audit trail. Its website is operated by Price-Jones Partners Ltd, London. Its live demo requires a verified business email. The table maps the vendor's stated capabilities to Delper's processes. These are the vendor's own claims and have not been verified.

Limits: The Envoy-based interception works at pod level in a cloud-native environment. It doesn't cover factory operational technology, which the Aegis hardware controls protect. Adopting SentraMesh would add a new provider and depend on Envoy and Kubernetes, so it would need approval, a contract that gives GK Bharta the kill switch, due diligence on Price-Jones Partners, and a place in the responsibility matrix.

Alternative: Forcepoint AI Agent Gateway

Forcepoint AI Agent Gateway is a second option for the same control. It sits between agents and the applications they call, and enforces what agents can reach, what data they can read, and which actions need a human to approve them. Its published claims are compared with SentraMesh below. These are the vendor's own claims and have not been verified.

Fit with Delper: The regulatory mapping covers DORA and NIS2, which matter for the EU markets and MiCA work in the plan. Forcepoint's credential brokering would remove standing credentials from the partner and SC integrations (PH 06, PH 19). Neither vendor's material mentions factory operational technology, so Aegis hardware controls still protect the factories. Only one of the two should be adopted. Each would need approval, a contract giving GK Bharta the kill switch, and a place in the responsibility matrix.

Voice and chat agent guardrails (ElevenLabs Guardrails 2.0)

ElevenLabs Guardrails 2.0 is a runtime control for ElevenAgents, its voice and conversational agent product. It checks each agent response against policies before the response reaches the user. The table maps its published claims to Delper. These are the vendor's own claims and have not been verified.

Fit and limits: This control suits voice and chat agents that talk to partners, distributors, merchants and consumers, such as the AI chatbot in PH 06 and the onboarding calls in PH 10 to 16. It does not cover back-office agents, settlement agents or factory systems, which the SentraMesh or Forcepoint options and the Aegis controls address. Adopting it would add a new provider, so it would need approval, a contract giving GK Bharta the kill switch, and a place in the responsibility matrix. Its cost is not stated on the page.

Business process reconciliation agent (agent 10)

Agent 10 is an accounting and reconciliation agent. It checks that every money and record flow in the business process agrees with its source. It works read-only: it matches, flags and reports, but it cannot post entries, release payments or change balances. GK Bharta finance keeps the ledger and all payments, so the agent adds a check, not a control over money.

Escalation: A break above the tolerance set by GK Bharta finance goes to a named owner the same day, and the agent records it in the audit trail. The agent does not correct the break itself. It also produces the evidence pack for EY's audit and PwC's compliance work.

Agent 10 is a proposal. Its tolerances, sources and go-live month are for GK Bharta finance to set.

AI agent register

The register lists the nine AI agents in the Delper business process, with scope, data access, accountable owner and authority. The agents are a proposal derived from the Launch Master Plan and Section 14, for the Group to confirm. Each agent operates within parameters, every decision outside them goes to a named human owner, and GK Bharta can halt any agent through the kill switch. No agent holds payment authority, since GK Bharta alone operates finance and payments.

AI build partner (Chizl)

Chizl is a Dubai-based technology company (Chizl Technology Solutions FZ-LLC) that builds forward-deployed AI platforms on the client's own infrastructure. It separates product, model and platform work into three workstreams, and states that it works in regulated sectors including financial institutions and government. It could build the nine agents and their interfaces, rather than the Group building them in-house. These are the company's own claims and have not been verified.

Role limits: Chizl would be a build and platform partner only. It must not operate finance or payments, run compliance, or perform the financial audit, since those roles sit with GK Bharta, PwC and EY. Adopting it would need approval, a contract giving GK Bharta the kill switch over its work, and a place in the responsibility matrix. Its sector list does not include supply-chain, token or payments work, so that experience should be checked during due diligence.

Payment acceptance processor (Adyen, option)

Adyen is a global payments platform that accepts card, wallet and local payments and operates in India under Reserve Bank of India authorisations. The plan names Stripe, Adyen and Worldpay as payment rails for the POS layer. The table maps Adyen's published claims to Delper. These are the company's own claims and have not been verified.

Finance rule: Adyen would handle incoming customer payments only. It would hold no authority over outgoing payments. GK Bharta alone operates finance and makes all outgoing payments, with a 2-day lead time, and this is unchanged. Any settlement from Adyen to Delper should pass through GK Bharta finance. Adopting Adyen would need approval, a contract that gives GK Bharta the kill switch over its services, and a place in the responsibility matrix. Its price is not stated on the page.

Channel connectors

Customers, vendors and partners reach Delper through the channels below. Each channel is a connector: it moves data and requests between the platform and the nine agents, and it does not take over any agent's authority. Connection, data use and platform terms need review before any channel is switched on. Outgoing payments stay with GK Bharta finance under the 2-day rule. Channels with no agent in the register are shown as a gap.

Product master data (PiLog Article Master, option)

PiLog Article Master is a retail product-data platform. It uses AI matching to flag duplicate product records, applies governance workflows before records reach SAP Retail, and builds one golden article record per product. Its use case is retail, so it fits the supplier and merchandise data in the supply-chain and channel work rather than the token or service flows. The table maps its claims to Delper. These are the vendor's own claims and have not been verified.

Limits: The matching engine only recommends. Final approval stays with a human steward, consistent with Section 12. Adopting it would add a new provider, so it would need approval, a contract that gives GK Bharta the kill switch over its work, and a place in the responsibility matrix. Its price is not stated.

Compliance-first go-live gate

No agent goes live in a market until that market's compliance conditions are signed off. The gate applies to each agent in each market, and the earliest go-live month shown below is a floor, not a start date. The licence programme (VARA, MAS, FCA, DFSA, FinCEN, NY BitLicense and MiCA) sets the market-by-market timing. Sign-off sits with the named owner, and GK Bharta can halt any agent at any time through the kill switch.

Build cost (internal estimates)

Internal build cost for the costed tasks in the build schedule. Each figure is team size × months × $5,000 per FTE-month, which is the $60,000 loaded annual cost. Team sizes are assumptions. Spend is recognised evenly across each task's window. Milestones and waves are excluded here.

Task detail

Each task is listed with its phase, number and category. Categories follow the plan's tags.

PH 01: Incorporation & Corporate Structure (M1–2, 10 tasks)

PH 02: Legal, Regulatory & Licensing (M1–20, 16 tasks)

PH 03: Banking, Treasury & Financial Operations (M1–4, 16 tasks)

PH 04: Executive Hiring & HR Infrastructure (M1–6, 17 tasks)

PH 05: Core Blockchain & Smart Contract Infrastructure (M1–6, 15 tasks)

PH 06: Product, Apps & Platform Development (M2–8, 20 tasks)

PH 07: Quality Assurance, Testing & Security (M4–8, 12 tasks)

PH 08: Token Economics Design & Minting (M1–4, 16 tasks)

PH 09: Exchange Listings & Market Liquidity (M8–18, 12 tasks)

PH 10: Opinion Poll → Consumer Onboarding (Layer 1) (M4–8, 12 tasks)

PH 11: Distributor Onboarding — Layer 2 (×0.850) (M5–10, 9 tasks)

PH 12: Manufacturer Onboarding — Layer 3 (×0.723) (M6–11, 9 tasks)