About Me
Senior Data Scientist / Applied AI Engineer with 5+ years building production AI — NLP pipelines, RAG systems, healthcare AI. The question I keep coming back to is how do you make this actually work in production, not just in a notebook.
My Story: Building Production AI Systems
My M.Tech in Data Science at IIT Jammu taught me to think rigorously about evidence. At nference.ai, I built production NLP and multimodal AI for healthcare. What I learned: a model that's technically impressive but slow, expensive, or unreliable in production isn't actually done.
At Lamipak, I designed and built LamiOps (RAG-powered assistant, 10K+ daily users) and LamiTracker (200+ sources, 18+ countries) — architecture, retrieval design, evaluation frameworks, and the backend underneath all of it. That's the work I want to keep doing at a senior technical level: owning hard AI systems end to end, not handing the interesting parts to someone else.
My Journey: Data Science Foundations → Production AI Systems → Senior AI/ML Engineering
From rigorous data science foundations to owning production AI systems end to end, at increasing scale and complexity.
Data Science Foundations
M.Tech in Data Science, IIT Jammu (CGPA: 8.70/10). Built deep foundations in NLP, Deep Learning, LLM, and GenAI. Mentored 100+ students as Teaching Assistant.
Production AI Systems
At nference.ai, built NLP pipelines for healthcare - knowledge distillation (8× smaller, 9× faster), clinical extraction with BioClinicalBERT, multimodal AI. Learned: technical impressiveness means nothing if users don't want it.
Owning Systems End-to-End
At Lamipak, built LamiOps (10K+ daily users) and LamiTracker (200+ sources, 18+ countries) - architecture, AI pipeline, backend, deployment. This is where I proved I could own a hard technical system solo, at production scale.
Impact & Achievements
10K+ Daily Users
LamiOps - RAG-powered enterprise assistant, owned end-to-end. Shipped with a quality eval framework (Groundedness, Hallucination Rate, MRR). Retrieval time: hours → seconds.
200+ Sources, 18+ Countries
LamiTracker - Regulatory Intelligence platform. LLM-based extraction, full AI + backend build. Reduced manual R&D tracking effort by 95%.
Evidence-Based Decisions
Design metrics that reveal what's actually happening, not just what looks good in a report. Published research (AAAI 2024, NEJM AI).
Full-Stack Technical Depth
Built the systems myself, end to end - model, retrieval, backend, deployment - so I know exactly what's feasible, what's costly, and where the real trade-offs are.
Problem-First Engineering
Good systems start from real constraints, not impressive technology for its own sake. This shaped everything from healthcare AI at nference to the architecture decisions at Lamipak.
Teaching & Mentorship
Teaching Assistant at IIT Jammu (2019–2021), mentoring 100+ students. Explaining clearly is as important as building well.
Beyond the Code
Built on hard work, powered by a touch of talent, sprinkled with humor, and fueled by infinite nerdiness.
Staying Active
When I'm not debugging models, you'll find me balancing code with cardio: gym sessions for strength, running for endurance, and cycling for exploring new trails. Movement keeps the mind sharp and the ideas flowing.
Fledgling Bookworm
A growing passion for reading keeps me curious beyond technical papers. Whether it's exploring new ideas, learning from diverse perspectives, or simply unwinding with a good story, books are becoming an essential part of my routine.
Grounded in Mindfulness
Meditation helps me stay centered amidst the fast-paced world of AI. It's not just about relaxation, it's about clarity, focus, and maintaining perspective when tackling complex problems.
One thing to remember about me: my humility isn't flattery, it's a value shaped by a humble upbringing.