I build systems that think — production ML pipelines, fine-tuned language models, clinical AI, and data platforms at scale.
Currently into embedding fine-tuning, neural machine translation with DSPy, agentic workflows, LLM-as-a-Judge evaluation, and AWS Bedrock engineering.
2025 – now Senior AI/ML Engineer Gracenote (Nielsen)
2023 – 2025 Data Architect & Scientist TCS → J&J MedTech
2022 – 2023 COE R&D Software Engineer TCS → J&J
2020 – 2022 Data Engineer → Lead TCS → J&J Vision
2022 – 2024 MTech Industrial AI IIT Madras
2016 – 2020 BTech Information Tech KGEC
| openrouter-mcp-multimodal | MCP server for OpenRouter — text + image via 200+ models |
| saaki | ICU mortality prediction for sepsis-AKI · MIMIC-IV · AUROC 0.80 |
| biogemma | Gemma 3 1B fine-tuned on medical corpus for clinical NLP |
| icd10cm | ICD-10-CM browser — 70K+ diagnosis codes · MongoDB + Node.js |
| Recurrent-Neural-Networks | Stock price prediction using LSTMs |
| GANs-using-pyTorch | GAN from scratch in PyTorch |
| HuggingFace Models | 8 fine-tuned models + 10 datasets |
5 peer-reviewed papers · Springer · IEEE · Elsevier
| Paper | Venue |
|---|---|
| FemmeBand: IoT Smart Security Band using EMG Sensors | Springer 2022 |
| HeartHealth: Multi-Attribute Heart Monitoring via Fuzzy-TOPSIS | IEEE 2021 |
| Predicting Bradycardia & AFib using Fuzzy Logic + IoT | Elsevier 2021 |
| ValveCare: Fuzzy-Based Heart Disease Prediction with IoT | Springer 2021 |
| SafeBand: IoT Smart Security Band with Instant SOS | Springer 2021 |
PyTorch Transformers scikit-learn LangChain LangGraph DSPy SHAP W&B AWS Bedrock Redshift PostgreSQL Snowflake Docker Terraform Python SQL Java Go Bash R
Lead guitarist 🎸 · Gamer 🎮