Summary
Varun Tulsian is an AI engineer with a decade of experience building production-grade ML systems that prioritize measurable outcomes and user-facing reliability. He led 0→1 GenAI and retrieval efforts at Needl.ai, shipping long-document finance QA and custom reporting tooling that turned 1–2.5k page PDFs into analyst-ready artifacts with citations and trust scores. At Walmart he connected ranking and personalization work to retail KPIs across marketplace and search, and earlier built a real-time personalization engine handling 10M+ events/day. Varun favors an eval-first, small-scope shipping cadence—thin slice, measure, then scale—and enforces cost guardrails, eval harnesses, and clear ownership across retrieval/eval/serving/UI. He combines applied ML research instincts with hands-on deployment experience (layout-aware parsing, bi-/cross-encoders, on-prem LLM serving), and has grown teams while keeping a strict stance against silent errors in user-facing flows. Based in Bengaluru, he’s practical, evidence-driven, and repeatedly turns complex long-doc and search problems into auditable, high-precision products.
10 years of coding experience
9 years of employment as a software developer
Master of Engineering (MEng) Computer Science, Master of Engineering (MEng) Computer Science at Indian Institute of Science (IISc)
10th, 10th at S. D. Nopany Vidyalaya
Bachelor of Technology (BTech) Information Technology, Bachelor of Technology (BTech) Information Technology at Heritage Inst. of Tech. Kolkata
12th, 12th at CBSE Birla High School
English, Hindi, Bengali