https://github.com/julientoucoula17/wallet17
Guided Asset Portfolio Allocation Platform with Machine Learning 📈
https://github.com/julientoucoula17/wallet17
flask mlflow python yfinance-api
Last synced: about 1 month ago
JSON representation
Guided Asset Portfolio Allocation Platform with Machine Learning 📈
- Host: GitHub
- URL: https://github.com/julientoucoula17/wallet17
- Owner: julientoucoula17
- Created: 2022-04-19T12:51:19.000Z (about 4 years ago)
- Default Branch: main
- Last Pushed: 2026-02-27T00:41:04.000Z (3 months ago)
- Last Synced: 2026-02-27T07:43:46.227Z (3 months ago)
- Topics: flask, mlflow, python, yfinance-api
- Language: HTML
- Homepage: https://julientoucoula17.github.io/wallet17/
- Size: 7.81 KB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
Wallet17 - Guided Asset Portfolio Allocation Platform with Machine Learning
## Context
Developed a scalable web application for guided portfolio allocation using deep learning and optimization to maximize risk-adjusted returns. Designed for finance, banking, and insurance with compliance, auditability, and operational efficiency.
## Objective
Automate and enhance asset allocation strategies by integrating real-time market data, LSTM predictive modeling, and a secure interface tailored to investor profiles.
## Business Value
- Improved decision-making via historical trends + ML forecasts.
- Reduced allocation/rebalancing time and operational risk.
- Compliance-ready with full traceability of decisions and changes.
- Adaptable to wealth management and insurance investment strategies.
## Technical Stack
- Backend & API: Python 3, Flask (RESTful)
- ML: LSTM forecasting, custom optimization engine
- Data Sources: yfinance, custom inference API
- Database: PostgreSQL, SQLAlchemy ORM, S3-compatible storage
- MLOps: MLflow for tracking, model versioning, reproducibility
- Deployment: Dockerized for scalable production
## Key Features
- Real-time data retrieval, preprocessing, and forecasting
- Allocation based on investor profile, forecasts, and risk optimization
- Role-based authentication, secure access, model version control
- End-to-end audit trail for allocations and decisions
## Role & Delivery
Agile team of four; contributed to ML pipeline (LSTM), backend integration, DB design, and deployment automation. Applied CI/CD for reliable releases.
## Team contributors
G. Richard, J. Jaewook, T. Steve, C. Samuel
## References
- https://www.kdnuggets.com/2019/06/optimization-python-money-risk.html
- https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2740027
- https://www.pragcap.com/
- https://disciplinefunds.com/what-is-the-discipline-fund/
- https://www.tensorflow.org/tutorials/structured_data/time_series