This repo gets better every time someone who actually does this work reads it and says "you missed something."
This is the only open-source resource that maps Indian financial services operations at the granular level needed to actually automate them. Every contribution makes it more accurate, more comprehensive, and more useful for the thousands of people who will use it to transform their organizations.
Your name goes on a resource that could reshape how 2-3 lakh ops professionals work.
If you work in settlement, KYC, compliance, commission processing, or any back-office function at a broker, AMC, distributor, or insurer -- we want your corrections. Every process we describe should match reality. If it does not, tell us.
If you have built an agent for any of the 52 use cases (even partially, even as a POC), we want your implementation learnings. What worked? What did not? What did the LLM get wrong?
SEBI circulars change quarterly. IRDAI guidelines are evolving rapidly. AMFI norms get updated. If you spot something outdated or incorrect in our regulatory sections, please fix it.
If you are building a product that automates any of these operations -- or if you have tried and learned why it is hard -- share your perspective. War stories are better than theory.
If you have benchmark data on LLM accuracy for financial document extraction, reconciliation, or classification -- we want it. Numbers beat claims.
- Star the repo (seriously, it helps with discoverability)
- Fork the repository
- Create a branch (git checkout -b feature/your-contribution)
- Make your changes following the guidelines below
- Submit a Pull Request with a clear description
For quick corrections: Just open an Issue. We will fix it.
For substantial additions: Please discuss in Issues first so we can align on scope and approach.
Be specific to Indian BFSI. Every example should use Indian system names (CAMS not Pershing, CDSL not DTCC, SEBI not SEC, BSE StAR MF not Schwab).
Be practical, not theoretical. If you describe a process, include: who does it, what system they use, what data they handle, how long it takes, what goes wrong.
Write for three audiences simultaneously:
- Ops person with 20 years experience should nod and think "yes, that is exactly how it works"
- PM with no BFSI background should understand what the process is and why it matters
- Developer who needs to build it should have enough detail to start coding
All contributors will be acknowledged in the README. Significant contributions will be credited in the relevant chapter.
Thank you for making this resource better. The next person who reads this might be your colleague, your competitor, or your future hire. Let us make sure they find something excellent.
This repo gets better every time someone who actually does this work reads it and says "you missed something."
This is the only open-source resource that maps Indian financial services operations at the granular level needed to actually automate them.
Your name goes on a resource that could reshape how 2-3 lakh ops professionals work.
If you work in settlement, KYC, compliance, commission processing, or any back-office function at a broker, AMC, distributor, or insurer -- we want your corrections.
If you have built an agent for any of the 52 use cases (even partially, even as a POC), we want your implementation learnings.
SEBI circulars change quarterly. IRDAI guidelines are evolving rapidly. AMFI norms get updated. If you spot something outdated or incorrect in our regulatory sections, please fix it.
If you are building a product that automates any of these operations -- or if you have tried and learned why it is hard -- share your perspective. War stories are better than theory.
If you have benchmark data on LLM accuracy for financial document extraction, reconciliation, or classification -- we want it. Numbers beat claims.
- Star the repo (seriously, it helps with discoverability)
- Fork the repository
- Create a branch (git checkout -b feature/your-contribution)
- Make your changes following the guidelines below
- Submit a Pull Request with a clear description
For quick corrections: Just open an Issue. We will fix it.
For substantial additions: Please discuss in Issues first so we can align on scope and approach.
Be specific to Indian BFSI. Every example should use Indian system names (CAMS not Pershing, CDSL not DTCC, SEBI not SEC, BSE StAR MF not Schwab).
Be practical, not theoretical. If you describe a process, include: who does it, what system they use, what data they handle, how long it takes, what goes wrong.
Write for three audiences simultaneously:
- Ops person with 20 years experience should nod and think "yes, that is exactly how it works"
- PM with no BFSI background should understand what the process is and why it matters
- Developer who needs to build it should have enough detail to start coding
All contributors will be acknowledged in the README. Significant contributions will be credited in the relevant chapter.
Thank you for making this resource better.