We will be undergoing planned maintenance on January 16th, 2026 at 1:00pm UTC. Please make sure to save your work.

Inspiration

Sustainability-linked loans with ESG are becoming increasingly in mainstream, as capital markets are rapidly linking the cost of debt to ESG performance. However, manual ESG due diligence, fragmented data across counterparties, and inconsistent KPI standards create operational bottlenecks in origination, covenant-linked pricing, monitoring, and reporting. My platform "GLC" brings together consistent ESG data, automates KPI-based credit assessments and covenant enforcement, and provides an auditable borrower-lender workflow delivering transparent SLL management that meets LMA ESG Frameworks, industry's need. It could improve efficiency and speed by 30%.

What it does

The "GLC Platform" provides a uniform platform for both borrowers and lenders, incorporating a compliance/monitoring framework that applies from loan origination throughout the entire life cycle of Green Loan Principles (GLP) and the LMA framework. It also offers services such as loan data resolution, loan assessment, sustainability checks, key stakeholder evaluation, loan balance sheet generation, and AI-assisted advice. This helps both borrowers and lenders reduce pressure, streamline processes, improve communication, and maintain compliance + reporting.

GLC Platform developed to serve the needs of:

  • Digitise the LMA Green Loan Principles workflow
  • Automate ESG metric extraction from borrower documents
  • Standardise compliance scoring across all applications
  • Scale sustainable finance without scaling headcount

KEY FEATURES:

Target Users:

For Borrowers

Application Assessment Form: Comprehensive loan form with ESG compliance questionnaires and Documents Upload
Loan Assets Management: Tracking and organizing loan documents with LOAN_ID and decentralized sharing across different shareholders for quick access
Invite Collaborator and Shareholder: Demo shareholders invitation with link
Loan Audit" View only access to audit pages for transparency
Real-time Status: Track application progress through review stages
User Guide: Guide and doucmentation to use platform
Learning Resources: SSL, ESG, and LMA resources to read to promote knowledge

For Lenders

Application Assessment Form: Same comprehensive loan form with ESG compliance questionnaires and Documents Upload for lenders so that lenders can verify any loan from their end
Loan Assets Management: Same tracking and organizing loan documents with LOAN_ID and decentralized sharing across different shareholders for quick access
Invite Collaborator and Shareholder: Same demo shareholders invitation with link
Dashboard: To see all loans summary and quick overview
Loan Audit Full control to check loan assessment and analysis, loan cycle management and decision-taking capabilities
GLP Compliance Check: 4 core components validated automatically
DNSH Assessment: 6 EU Taxonomy criteria evaluated
Carbon Lock-in Risk: Stranded asset risk identification
AI Document Chat: Ask questions about uploaded sustainability reports
Location Intelligence: Project location mapping with environmental data
User Guide: Guide and doucmentation to use platform
Learning Resources: SSL, ESG, and LMA resources to read to promote knowledge

How I built it

I started by exploring the LMA documents and resources at 'lma.eu.com/sustainable-lending/resources' and the LinkedIn community to ensure the development adhered to industry standards. The hackathon version's architecture relies on open-source models, APIs, and libraries.

The GLC platform is built with a backend-first approach, utilising FastAPI for the backend, JavaScript for the frontend, Hugging Face and Langchain for AI tasks, as well as document and data processing libraries and open-source sustainable and environmental data APIs.

GLC Platform Architecture – ESG-Linked Loan Management System

Tech Stack

Layer Technology Purpose
Backend Python 3.12 + FastAPI High-performance async API
Database SQLite + SQLAlchemy Lightweight, portable storage
Vector Search FAISS + Vector DB + MiniLM-L6-v2 Semantic document search
NLP/AI RoBERTa (QA) + Flan-T5 (RAG) Document understanding
Frontend Vanilla JS + TailwindCSS Fast, responsive UI
APIs Hugging Face and Open-meteo For AI Models and Environmental Data
PDF Processing pdfminer.six + PyPDF2 Text extraction and export

Challenges I ran into

  • Create an effective loan assessment form that collects only the necessary information for faster processing.
  • Gathering, retrieving, and calculating key metrics
  • Creating an AI RAG system with a compatible direct inference and maintaining sufficient context length.

Accomplishments that I'm proud of

  • Direct alignment with LMA GLP Framework and ESG (compliance + reporting).
  • Scalable & modular: API-first architecture means banks can plug it into existing stacks (agent banks, syndication, fund onboarding).
  • Automatic calculation of high-impact metrics and ESG-principal compliance check: time saved in diligence, reduction in manual errors, faster deal execution - all judgeable.

What I learned

  1. Importance of Growing Sustainability-linked loans
  2. Role of different parties involved in a loan, such as borrower, lenders, investors, agents, regulators, ESG, and civil society and how all collaborate together on ESG truth.
  3. Capturing and verifying ESG data across different areas and making it distributed and accessible along with its syndication.
  4. How to serve an AI agent with direct inference?

What's next for GLP Cycle

  1. Automatic loan assessment filing by uploading the company's sustainability-linked report document and annual report.
  2. Collaborate with the loan market professional and ESG reviewers to test and scale the platform for real-world tasks.
  3. Connect it with existing tools and solutions to enhance functionality and load balance.
  4. Add a robust finance agent to research the company's profile and provide reliable financial information.

Thank You!
Below is the link to 'pitch deck' : Canva

Built With

Share this project:

Updates