Welcome to my GitHub! I'm a passionate AI researcher and engineer, constantly exploring the frontiers of machine learning, NLP, and multimodal AI. My work spans across academia and industry, where I develop scalable ML algorithms, optimize NLP pipelines, and contribute to cutting-edge advancements in AI.
- PhD in Computer Engineering (Stony Brook University) | Research: Scalable ML, Deep Learning, Reinforcement Learning, NLP
- Published at CAI 2025, SSCI 2025, CCN 2024, CITDS 2024, SMACD 2024 β on topics like LLMs, Semantic Classification, Speech-driven AI, and IoT optimization.
- Researching SLU methodologies, reinforcement learning for team optimization, and knowledge graph-driven AI.
- Deep Learning, LLMs, NLP, Speech Recognition, Reinforcement Learning
- Finetuning: LoRA, QLoRA | RL Techniques: RLHF, PPO, DPO
- Libraries: PyTorch, TensorFlow, Hugging Face, Langchain, OpenCV
- Python, C++, JavaScript
- MLOps & DevOps: Git, Docker, MLflow, DVC
- Databases: Vector DBs (Qdrant, Pinecone), Neo4j, Redis, Firebase
- Mobile Apps: React Native, Flutter
- π Built GraphRAG-based retrieval model for structured knowledge understanding
- π Processed 300+ pages on Stereo Vision, boosting retrieval accuracy to 95%
- ποΈ Achieved 2s response time, reducing study time by 25%
- π£οΈ Real-time translation for call centers, supporting 10+ languages
- β‘ 40% lower latency, Dockerized deployment for scalable accessibility
- π€ Automated agent conversation analysis with NLP (1,000+ daily interactions)
- π― Achieved 85% accuracy in soft skill trait classification
- Exploring multi-agent reinforcement learning for dynamic team optimization π€
- Graph-based AI retrieval for knowledge-driven applications π
- LLM interpretability & causal tracing to enhance model decision-making π
- LinkedIn: linkedin.com/in/villurignanesh
- Email: villurignanesh@gmail.com
Let's collaborate and build the future of AI! π