Headline

LLM Inference | Agentic AI | RAG | AI Software Engineer, DataCenter AI Software @Intel

About

AI Software Engineer with 4+ years of experience specializing in LLM inference optimization, agentic AI systems, and MLOps at scale. I build systems that make large language models fast enough, cheap enough, and reliable enough to actually ship to customers. For the past two years at Intel, that's meant leading deployment engineering for our AI for Enterprise Inference platform — an open-source stack that took LLM deployment from a multi-day process down to about 30 minutes, and it's now running in production for 10+ enterprise customers. A lot of my day-to-day comes down to a simple question: how do you take a model that works in a notebook and make it work under real load, real latency budgets, and real cost constraints? FP8 quantization got us 2x memory savings and 1.6x more throughput without hurting accuracy. Disaggregated prefill/decode serving with vLLM and LLM-d cut time-to-first-token by 63%. None of that is glamorous work, but it's the difference between a demo and a product. Before this, I spent two years building the Kubeflow platform that 100+ data scientists relied on daily, and led a migration that made our logging infrastructure 3x more scalable while cutting costs by 40%. I'm at my best when I'm the person translating "cutting-edge research" into "actually works in production, at scale, for real users" — and I'm increasingly curious about applying that same discipline to more hardware-constrained environments, like edge and embedded AI, where the tradeoffs get even sharper. Always happy to talk inference optimization, agentic systems, or how to make LLMs boringly reliable at scale.

Location

Bengaluru

AI stack

claude code

claude

github copilot

hugging face

Work experience

AI Software Engineer at Intel Corporation, Nov 2024 - Present · 1 yr 10 mos, Bengaluru

Software Engineer at Intel Corporation, Jun 2022 - Nov 2024 · 2 yrs 6 mos, Bengaluru, Karnataka, India

Graduate Intern Technical at Intel Corporation, Jul 2021 - May 2022 · 11 mos, Bangalore Urban, Karnataka, India

Education

Master of Technology - MTech, Computer Science and Engineering, National Institute of Technology Durgapur

Bachelor of Technology - BTech, Computer Science and Engineering, Bikaner Technical University | University College Of Engineering & Technology