A production AI platform I architected end to end, plus 14+ open-source projects across RAG, machine learning, NLP, infra, and web, plus a stack of Kaggle notebooks.
My biggest project lives in production, not on GitHub. Here's what it does.
A production AI platform for commercial real estate lending whose core I built, turning raw financial documents into auditable credit decisions. Multi-LLM orchestration, a RAG document-intelligence pipeline, and a deterministic credit risk engine, deployed to AWS.
Classification + extraction over 24 CRE document types, OCR routing, dedup, and pgvector retrieval with reranking.
Deterministic, rule-based scoring (DSCR, LTV, NOI, FICO) with tiered decisions and zero LLM in the critical path.
Provider fallback across Bedrock, Gemini & OpenAI with per-provider circuit breakers and graceful degradation.
End-to-end projects, with source on GitHub unless noted
github: @bijay-odyssey / visibility: public
Python · scikit-learn · pandas · PyPI · CI
An open-source Python library for leakage-safe EDA and ML preprocessing, built around an explainable planner. It inspects a dataset, works out which preprocessing operations actually apply, states why in plain language, and fits every statistic on the training split alone.
explain() reports the statistic behind each decision: median imputation
because skew is 3.22, target encoding with 5-fold cross-fitting because 163 levels
exceeds the one-hot ceiling, an identifier column dropped because 100% of its values
are distinctPython · Qdrant · open source
Formula evaluation in Qdrant's official Python client accepted zero raised to a negative exponent, which is undefined, instead of rejecting it. Reviewed and merged by a Qdrant maintainer after CI passed.
FastAPI · pgvector · Arq · Tesseract · React · source private
A final-year college project and an end-to-end applied AI system for Nepal's SMEs: it ingests tax documents, extracts and classifies them, flags anomalies, and produces IRD-ready VAT and TDS reports. Sole developer of the system, in a two-person team.
Python · FastAPI · Pydantic v2 · pytest
An orchestration layer over a fleet of local services. It routes a natural-language request to whichever project can answer it, starts that service on demand, and says plainly when the capability does not exist instead of improvising one.
Python · Apify · pandas
A scraping and analysis pipeline for hiring-market data, built to replace opinion about in-demand skills with measured evidence.
Python · CLI · MIT
A debate engine that argues the opposite of whatever you claim, without ever fabricating to do it. Useful as a devil's-advocate pass over an argument you are too close to.
GitHubPython · FastAPI · PostgreSQL · pip package · CLI
A hierarchical task-orchestration framework for a solo developer running many small services: a shared registry and a mission board that independent workers ("disciples") poll, claim atomically, and report back to. One HTTP service, one Postgres database, two tables, no broker and no queue, sized to run on a free tier.
UPDATE, with no read-then-write race
window, verified by a test that fires 20 concurrent claims and asserts exactly one winnerFastAPI · FAISS · Qdrant · Groq · Llama-3.3-70B
Modular RAG API built from scratch with FastAPI for document upload and natural language querying. Features switchable vector backends (FAISS/Qdrant), multiple chunking strategies, cross-encoder reranking, and Groq-powered generation with Llama-3.3-70B.
FAISS · Cross-Encoder · Reranking
Comparative study showing why retrieval alone isn't enough. Two-stage retrieval + reranking pipeline with noise injection to stress-test semantic search performance.
Flask · Scikit-learn · SHAP · SQLite
Flask web app for real-time churn prediction with user authentication, admin dashboard, CSV export, and SHAP-powered explainability. Automated retention strategy generator included.
XGBoost · SHAP · Imbalanced Learning · Optuna
Transaction classification on 1M+ Kaggle IEEE-CIS records. Achieved ROC-AUC 0.95 and F1 0.66 on the imbalanced fraud class with Optuna hyperparameter tuning.
LightGBM · Prophet · Feature Engineering
SKU-level supermarket price forecasting on 1.7M+ rows using LightGBM and Prophet with lag, rolling window, and calendar features.
Random Forest · Optuna · SHAP
Competition-aware ML workflow integrating historical sales and competitor pricing. Achieved R² ≈ 0.92 with Random Forest and SHAP explainability for business decisions.
Isolation Forest · LOF · KMeans · DBSCAN
Multi-method anomaly detection pipeline combining model-based, cluster-based, and statistical approaches on fraud and retail datasets with PCA visualization.
sentence-transformers · FAISS · Visualization
Scripts and experiments for exploring text embedding models, visualizing embedding spaces, and benchmarking retrieval quality across different encoder models.
GitHubFAISS · sentence-transformers · Python
End-to-end semantic search implementation using FAISS for fast approximate nearest neighbor search with sentence-transformers for dense embedding generation.
GitHubQdrant · Python · Vector Search
Experiments with Qdrant vector database: collection management, payload filtering, hybrid search, and performance benchmarks for RAG applications.
GitHubPython · langchain-text-splitters · Benchmarking
Comparative study of text chunking strategies (fixed-size, sentence-based, recursive, and semantic chunking), evaluating their impact on RAG retrieval quality.
GitHubScikit-learn · XGBoost · LightGBM · Optuna
Collection of regression projects across different domains (housing prices, demand forecasting, and insurance costs), with thorough EDA, feature engineering, and evaluation.
KMeans · DBSCAN · PCA · Silhouette Analysis
Unsupervised learning project exploring clustering algorithms for customer segmentation. Includes PCA for dimensionality reduction and silhouette analysis for optimal cluster selection.
No projects found for this filter.
27 public notebooks · 58 upvotes: community-tested data science