AI Systems Engineer @ Benevity | GenAI & ML Engineer | Data Scientist
Automating today so we can innovate tomorrow.
I’m an AI Systems Engineer with a background in Generative AI, Machine Learning, and Data Science, building enterprise-grade AI solutions that scale.
I specialize in Agentic AI workflows, RAG systems, and ML pipelines — bridging raw data chaos into business-ready automation.
- 🎯 Current: Building Agentic GenAI systems at Benevity to power CSR, ESG, and employee engagement.
- 🏎 Past @ Honda: Deployed ML models for Warranty Claim Classification, PDCA AI Summaries, and MDM Customer Matching.
- 🌍 Experience: From healthcare AI at BOH Pharma to enterprise AI systems at LTIMindtree & Honda.
GenAI & LLMs:
LangChain • LangGraph • Ollama • Azure OpenAI • Watsonx.ai • Llama • Gemma • RAG • LLMOps
ML Engineering:
Python • scikit-learn • XGBoost • TensorFlow • PyTorch • Spark • Databricks
MLOps & DevOps:
GitHub Actions • Jenkins • Docker • CI/CD • Watson Studio • OpenScale
Data & Analytics:
BigQuery • MongoDB • MySQL • Pandas • Power BI • Tableau
- Agentic PDCA Assistant → Conversational RAG app summarizing 200+ project decks weekly for executives.
- Warranty Claim Classification → Hybrid ML + GenAI model reducing QE workload by 10–20 hours/day.
- DevOps Release Notes Agent → Multi-agent LLM workflow generating audience-specific release notes.
- MDM Customer Matching → Classification model with 99% precision, saving 25% manual work.
- Healthcare AI (BOH Pharma) → Built ML models for early cancer detection with 92% precision.
Always building. Always learning. Always keeping it real.
Let’s talk about GenAI, Agentic AI, or real-world AI engineering.


