The Insight
LMA has built world-class transition finance standards. But they're written by lenders, for lenders.
The borrowers who need them most — African project developers facing a $233 billion annual financing gap — can't access them. A solar developer in Nairobi or a wind farm operator in Lagos will never navigate thousands of technical PDFs written for global banks and law firms.
The knowledge exists. The standards exist. The projects exist. The capital exists.
What's missing is the bridge.
The Opportunity
| The Gap | The Resource | The Disconnect |
|---|---|---|
| $233B annual climate finance gap in Africa | 2,009 LMA documents (2013-2025) | Written for lenders, inaccessible to borrowers |
| 60% of world's best solar resources | 83 Africa-specific LMA materials | Scattered across technical PDFs |
| $700B+ global transition lending market | October 2025 Transition Loan Guide | No tool to operationalize it for developers |
Why now: The October 2025 LMA Transition Loan Guide finally introduced clear, unified criteria for credible transition finance. For the first time, there's a standard that can be encoded, automated, and made accessible.
What Verdex Does
Verdex is the borrower's toolkit for LMA transition finance compliance.
We help project developers understand what lenders need — before they walk into the room.
| Feature | What It Does | Why It Matters |
|---|---|---|
| LMA Transition Validator | Scores projects against the 5 Core Components | Developers know if they qualify before approaching lenders |
| EU Taxonomy DNSH | Screens against 6 environmental harm criteria | European DFIs require this — now developers can self-assess |
| Climate Intelligence | Location-specific climate data and 2050 projections | Proves project is designed for local conditions |
| Clause Search + AI Ask | Semantic search across 500+ LMA clause templates | Developers understand the language lenders use |
| Greenwashing Detection | Flags red flags before they reach a lender's desk | Protects both sides from reputational risk |
| DFI Matching | Matches projects with appropriate financiers | Developers find the right capital, not just any capital |
How It Works
UPLOAD VALIDATE MATCH GENERATE
─────── ──────── ───── ────────
Project PDF → LMA 5 Core → 7+ DFIs → KPIs, SPTs,
or form + DNSH (6 criteria) ranked by documentation
+ Greenwash eligibility package
+ Climate Risk
⏱️ Under 60 seconds
| Step | Input | Output |
|---|---|---|
| 1. Upload | Project PDF or manual form | Structured project data |
| 2. Validate | AI analysis against LMA + DNSH criteria | LMA score (0-100) + DNSH score (0-100) + gap analysis |
| 3. Match | Project profile vs. DFI requirements | Ranked recommendations (IFC, AfDB, FMO, DEG, BII, Proparco, DFC) |
| 4. Generate | Validated project data | KPI framework + LMA-compliant documentation |
The difference:
| Traditional Process | Verdex |
|---|---|
| 2-4 weeks | < 60 seconds |
| $50,000+ consulting fees | Start for FREE! |
| Relationship-dependent | Criteria-based |
| Per-deal customization | Standardized framework |
Who Benefits
| Stakeholder | Their Problem | Verdex Value |
|---|---|---|
| Project Developers | Have a project and a PDF, lack expertise to structure bankable proposals | Self-assess against LMA standards, understand what lenders need |
| Commercial Banks | Can't verify transition claims, greenwashing liability risk | Pre-validated credentials, reduced due diligence burden |
| DFIs | No pre-screened pipeline, high origination costs | LMA-compliant deal flow, standardized assessment |
| LMA Ecosystem | Standards exist but adoption limited to lender side | Extended reach to borrowers, more deals using LMA frameworks |
The Foundation We Built
Knowledge Base
We didn't fabricate criteria. We read the documents.
| Category | Documents Catalogued | Indexed (MVP) | Source |
|---|---|---|---|
| LMA Total Library | 2,009 | — | lma.eu.com |
| Africa-Specific | 83 | 3 | OHADA frameworks, ZARONIA templates, regional guides |
| Sustainability & Green Finance | 161 | 13 | Green Loan Principles, ESG guidance, margin ratchets |
| Transition Finance | 42 | 3 | October 2025 Transition Loan Guide, LIBOR transition |
| External Frameworks | 2 | 2 | SBTi Net-Zero V2, Paris Agreement |
What we found: 12 years of LMA Africa engagement buried in conference reports, training materials, and legal guides that nobody had systematically indexed. These documents now power our clause search and AI assistant features.
Technical Architecture
| Component | Technology | Purpose |
|---|---|---|
| Validation Engine | Rules from October 2025 Guide + SBTi V2 | Encode real compliance criteria |
| DNSH Evaluator | EU Taxonomy Article 17 criteria | Screen 6 environmental objectives |
| Climate Intelligence | Open-Meteo + CMIP6 models | Location-specific risk and 2050 projections |
| Clause Search | Vector embeddings + semantic matching | Find relevant provisions, not just keywords |
| AI Responses | Source-cited answers only | Responses grounded in indexed LMA documents |
| Infrastructure | Cloud-native, stateless | Scales from one project to thousands |
Design Decisions
| Decision | Why |
|---|---|
| Scoped to verifiable data only | DFI database sourced from official websites. Eligibility criteria should be verified directly with each DFI. |
| Explicit about limitations | Clear distinction between validated vs. estimated, implemented vs. planned |
| LMA-specific, not generic ESG | Rules based on October 2025 Transition Loan Guide — the actual standard |
| Borrower-first interface | Simple enough for first-time developers, rigorous enough for DFI due diligence |
| DNSH assesses, not fabricates | Verdex identifies environmental gaps — it doesn't generate compliance that doesn't exist |
On DNSH Ethics
Verdex improves documentation, not fabricates compliance.
When the AI draft is re-uploaded, LMA scores improve because structure and methodology are added. But DNSH gaps remain if the underlying assessment was never conducted.
Example: If a project never mentioned water impact, the DNSH Water score stays at zero — because Verdex won't invent a water assessment that doesn't exist.
Why this matters: The worst thing we could do for African developers is help them submit documentation that doesn't reflect reality. When that project faces environmental issues later, it damages credibility for every African project that follows.
Verdex shows the path to compliance. It doesn't offer shortcuts around it.
Traction
| Metric | Status |
|---|---|
| MVP | Live at verdx.site |
| Documents catalogued | 2,009 |
| Africa-specific materials | 83 |
| Documents indexed (Phase 1) | 20 |
| Clause templates searchable | 500+ |
| DFIs in matching database | 7 |
| LMA principles covered | 100% of 5 Core Components |
Ecosystem Fit
Verdex complements, not competes with, existing LMA infrastructure.
Every tool in the loan market serves lenders. None serve borrowers.
┌─────────────────────────────────────────────────────────────────┐
│ │
│ LENDER SIDE BORROWER SIDE │
│ ─────────── ───────────── │
│ │
│ LMA.Automate (doc automation & negotiation) Verdex │
│ LendOS (loan servicing & portfolio) │
│ Agensight (agency workflow digitization) │
│ │
│ "How to draft, administer, "Do I qualify? │
│ and trade loans" What do lenders │
│ need from me?" │
│ │
│ ───────────────────────────── │
│ LMA STANDARDS │
│ (Same framework, both sides) │
│ │
└─────────────────────────────────────────────────────────────────┘
| Lender-Side Tools | What They Do | Verdex Difference |
|---|---|---|
| LMA.Automate (Avvoka) | Document automation & negotiation | We prepare borrowers before they reach you |
| LendOS | Loan servicing & portfolio management | We pre-validate before administration begins |
| Agensight | Agency workflow digitization | We standardize borrower submissions |
Value to lenders: Pre-validated borrowers = faster due diligence.
Value to LMA: More deals structured using LMA frameworks. Adoption grows on both sides.
Impact Potential
If Verdex enables just 10% of the financing gap to be addressed by 2030:
| Metric | Impact |
|---|---|
| Additional climate finance | $23 billion annually |
| Renewable capacity enabled | 50+ GW |
| People with electricity access | 100+ million |
| CO2e avoided | 100+ million tonnes annually |
Beyond climate outcomes: standardized LMA validation reduces friction across the industry. Every project assessed on the same framework means faster decisions, lower risk, and a common language between developers and lenders.
What's Next
| Today (MVP) | Phase 2 | Phase 3 |
|---|---|---|
| 20 docs indexed | 2,009 docs fully indexed | DFI pipeline integration |
| LMA 5 Components scoring | Template matching + deviation detection | Real-time deal flow |
| 7 DFI matching | Advanced filtering | Pan-African coverage |
| Semantic clause search | Lender-side dashboard | Industry-wide standardization |
Links
| Resource | URL |
|---|---|
| Live Demo | verdx.site |
| Documentation | verdx.site/docs |
| Pitch Deck | verdx.site/verdex-deck |
| Executive Summary | verdx.site/docs/executive-summary |
| Why Africa | verdx.site/docs/problem/why-africa |
| LMA Validator | verdx.site/docs/features/lma-validator |
| Knowledge Base | verdx.site/docs/technical/knowledge-base |
Verdex — Where African projects learn to speak the language of global lenders.
Built With
- asi:one
- fastapi
- groq
- next.js
- pinecone
- python
- railway
- react
- tailwind
- typescript
- vercel
Log in or sign up for Devpost to join the conversation.