My Projects

A selection of my work in agentic AI, RAG systems, LLM evaluation, NLP, computer vision, and quantitative analysis. Every project links to its source code on GitHub.

🔬

Deep Research Agent (Hybrid Multi-Agent)

Four LangGraph agents (Planner, Searcher, Synthesizer, Writer) produce citation-backed research reports. A local Qwen 3 model handles planning and synthesis while Gemini writes only the final report, cutting API requests ~74% and input tokens ~80%.

Python LangGraph LangChain Ollama Gemini Streamlit
Code
🕸️

Agentic RAG Engine with Verifier/Critic Loop

A reasoning agent routes sub-queries across a vector DB, SQL database, knowledge graph, and live web search, fuses the evidence, and runs a Verifier/Critic node that automatically retries low-confidence answers.

Python LangGraph ChromaDB SQLite NetworkX Streamlit
Code
📊

Agentic RAG Evaluation Harness

A benchmark suite that tests the Agentic RAG Engine against a single-pass RAG baseline using 24 ground-truth questions, an auditable LLM-as-judge rubric, and statistical analysis of groundedness, correctness, latency, and cost.

Python LLM-as-Judge Benchmarking RAG Evaluation Gemini
Code
📈

Quantitative Stock Analysis Pipeline

A multi-factor stock ranking pipeline that computes technical signals (RSI, MACD, EMA, ATR, OBV, VWAP), scores stocks across the universe, and measures signal uncertainty with Monte Carlo simulation and Bayesian updating.

Python pandas NumPy yfinance Monte Carlo Bayesian
Code
🚖

LLM-Driven Tabular Classification (NYC Taxi)

Can an LLM beat traditional ML on structured data? A quantized Mistral 7B classifies whether NYC taxi riders tip over 20% by reasoning over trip data converted into natural-language descriptions, reaching ~92% validation accuracy.

Python Mistral 7B Quantization Prompt Engineering Tabular ML
Code
⚖️

German-to-English Legal Translator

A domain-adapted machine translation model, fine-tuned from Helsinki-NLP's opus-mt-de-en on legal and university terminology so domain terms translate correctly (e.g. “Matrikelnummer” → “Student ID”).

Python Hugging Face PyTorch Fine-Tuning NLP
Code
🔒

Local RAG: Private CV Analyzer

A fully local RAG pipeline for chatting with PDF documents without sending data to the cloud, using Gemma 3 via Ollama, a Dockerized Weaviate vector DB, and OCR support for scanned files.

Python Ollama Gemma 3 Weaviate Docker OCR
Code
🎯

Auto-Labeling Pipeline for Object Detection

Generates training labels automatically with no manual annotation: a detection model finds objects, SAM 2 refines each one into a pixel mask, and geometric fitting turns the masks into YOLO oriented bounding boxes.

Python YOLO (Ultralytics) SAM 2 OpenCV Computer Vision
Code