machine-learning

Production-ready Python library for multi-provider LLM orchestration

An end-to-end image captioning project using a CNN encoder (ResNet-50) and LSTM decoder in PyTorch. Includes vocabulary building, preprocessing, training with BLEU evaluation, and inference. Generates natural language captions for images with saved metrics, model checkpoints, and visualization outputs.

ONNX-compatible Fast SeamlessM4T—Massively Multilingual & Multimodal Machine Translation

OpenLLM Monitor is a plug-and-play, real-time observability dashboard for monitoring and debugging LLM API calls across OpenAI, Ollama, OpenRouter, and more. Tracks tokens, latency, cost, retries, and lets you replay prompts — fully open-source and self-hostable.

AI Data Engineering Healthcare Platform: PySpark Medallion Lakehouse, Airflow DAGs, 5 ML Models, Local LLM RAG, and FastAPI Clinical Architecture.

Spark-Transformers: Library for exporting Apache Spark MLLIB models to use them in any Java application with no other dependencies.

Airflow Pipeline for Machine Learning

Interpretable and efficient predictors using pre-trained language models. Scikit-learn compatible.

ONNX implementation of the BGE-M3 multilingual embedding model and tokenizer with native C#, Java, and Python implementations. Generates all three embedding types: dense, sparse, and ColBERT vectors.

PyCrust is a CLI tool that converts Python code to Rust using OpenAI's GPT-3.5 language model for code generation. It provides an easy-to-use interface for converting Python code to Rust code, making it easier for Python developers to leverage Rust's performance benefits.

End-to-end pipeline for seeing how LLMs actually process your prompts. Capture attention across every layer, render heatmaps and cooking curves, compare variants with evidence — not vibes.

A python library for efficient KNN search within metric spaces using multiple distance functions.

[MobiCom 2022] InFi is a library for building input filters for resource-efficient inference.

[WACV 2025] Enhancing Scene Graph Generation with Hierarchical Relationships and Commonsense Knowledge

Tree prompting: easy-to-use scikit-learn interface for improved prompting.

ReAct AI Agent from Scratch using DeepSeek: Handling Memory & Tools without Frameworks

A simple and intuitive library for diffusion models using Flax and Jax. Includes detailed notebooks on DDPM, DDIM, and EDM with simplified mathematical explanations. Made as part of my journey for learning and experimenting with generative AI.

Modular and Universal AI

Run, quantize, and fine-tune LLMs on Apple Silicon. Pure Rust, no Python, no CUDA, no ONNX

Native AI inference for PHP 8.3+ - run ONNX, GGUF (llama.cpp) and RubixML models directly in your PHP process via FFI. Chat, streaming, embeddings, RAG and vector search. No Python, no HTTP microservices, no Docker sidecars.