Pinned Loading

1. Claim-Level-Halluciantion-Detection Claim-Level-Halluciantion-Detection Public

   A claim-level fact-checking system that uses RoBERTa to spot unsupported or contradictory scientific claims.

   Python

2. Multi-Source-Research-Analyst-Agent Multi-Source-Research-Analyst-Agent Public

   An LLM-powered AI agent that researches any question across the web, Wikipedia, and academic papers, then writes a clear, source-cited report. Built and automatically tested in Python.

   Python

3. Real-Time-Content-Moderation-and-Safety Real-Time-Content-Moderation-and-Safety Public

   Real-time multi-source content moderation pipeline for Bluesky and YouTube using Redpanda, Faust, Ollama, online topic clustering, Redis TimeSeries, and Streamlit.

   Python

4. End-to-End-MLOps-Pipeline-for-Fraud-Detection End-to-End-MLOps-Pipeline-for-Fraud-Detection Public

   Production-ready MLOps pipeline for real-time fraud detection with XGBoost, PyTorch, MLflow, FastAPI, drift monitoring, and Docker.

   Python

5. nim-multimodal-agent nim-multimodal-agent Public

   Multimodal agentic RAG on NVIDIA NIM: a LangGraph agent that routes retrieved figures through a Nemotron vision-language model and validates grounding with an LLM-as-judge benchmark.

   Python

6. RecallGraph RecallGraph Public

   Evidence-grounded recall tracing with LLM-assisted extraction, deterministic graph analysis, and auditable human review.

   Jac