deeploans is an open-source ETL framework designed for working with granular asset data. It extracts, cleans, transforms, and standardizes datasets, making them easier to use in analytics, machine learning models, and downstream applications.
Initially developed as an advanced credit risk analytics tool for portfolio managers and bankers, deeploans evolved into an open data infrastructure project to address the industry-wide challenge of inconsistent and fragmented financial data. It enables users to:
- Ingest loan-level datasets from multiple sources.
- Process and standardize data formats for easier integration.
- Validate and clean incomplete or inconsistent records.
- Output better-structured data suitable for analysis, modeling, data science or reporting.
deeploans is designed for analysts, data scientists, and developers working with granular asset data who need an efficient and cost-effective way to handle structured datasets without relying on third-party providers. It supports modern data workflows and removes the bottlenecks associated with manual data preparation.
deeploans includes ETLs for the following structured finance datasets:
- Auto Loans
- SME Loans
- Consumer Loans
- Residential Mortgages
- Commercial Mortgages
Additional components in this repository include:
api/api-backend-main/: FastAPI backend for data accessapp-library/: A library of (MVP) applications that run on top of deeploansmcp-server/: standalone MCP server for AI/client integrations
Deeploans is available under the Apache licence. See here for full text.
deeploans is a growing open-source project where developers and data analysts can directly shape the future of their tools and infrastructure.
🛠 Help build deeploans
We’re actively expanding deeploans, and now is the perfect time to jump in. Whether you’re experienced with ETL pipelines, data processing, or just getting started, now is the time to become a core part of the deeploans project.
🔨 Code contributions
Got an idea to improve our data processing? Spotted something that could use some tweaking? Even smaller contributions can make a real impact.
📊 Test and give feedback
Not a developer? No problem. Run Deeploans with real data, report bugs, and let us know how it performs. Your feedback makes the tool more robust for everyone.
Have an idea for improvement? Open a discussion—real-world testing is just as valuable as coding.
📖 Improve the docs
Documentation is everything. If something unclear or missing, your contributions to our docs can help the next developer get started faster.
💡Feature requests & ideas
Got a use case we haven’t covered? Open an issue or drop a comment in Discussions to brainstorm.
- Fork the repo and clone it locally.
- Explore the codebase and try it out.
- Have an idea? Open a discussion to talk about potential contributions.
- Once issues are posted, grab one and submit a pull request when you're ready.
- Stay tuned to join the community on Discord!
To get in touch, drop an email at:
deeploans began as an AI-powered analytics tool designed to help portfolio managers predict default risk.
However, we quickly recognised a fundamental challenge: before the market can adopt more advanced tools, it needs clean and better-structured data.
Thus, deeploans evolved into an open data infrastructure project. Today, deeploans solves the fragmented, inconsistent and incomplete data problem in finance, enabling analysts and developers to build better tools.