Professional Summary
Applied AI and software engineer at Guidewire, working across model evaluation, agentic systems, and production software. I build practical AI workflows with explicit quality checks and have delivered backend and full-stack systems using Python, Go, TypeScript, PostgreSQL, AWS, and Kubernetes. My experience also includes a deployed healthcare chatbot and published machine-learning research.
Current Focus
I evaluate where smaller, specialized language models can meet task requirements without defaulting to the largest model. The work spans open-model deployment, representative evaluation sets, LLM-as-a-Judge pipelines, observability, inference serving, and fine-tuning. I am particularly interested in reliable AI infrastructure: selecting models based on measured quality, cost, and latency, then integrating them into secure enterprise workflows.
Selected Proof Points
- Secure AI product engineering: Built secure natural-language-to-SQL capabilities for an enterprise product. View evidence
- Agentic AI recognition: Finalist at Google's Agentic AI Day. View evidence
- Applied AI engineering: Built this evidence-led portfolio and its grounded assistant. View evidence
- Published research: Co-authored work on data quality and water-body segmentation. View evidence
Education
B.Tech, Computer Science Engineering (Artificial Intelligence)
Amrita Vishwa Vidyapeetham, October 2020 - May 2024
CGPA: 9.12/10
Contact and Target Roles
Open to applied AI, ML engineering, AI platform, and agentic systems roles where evaluation and production engineering matter. Email me, connect on LinkedIn, or review my work on GitHub.