Thillai Chithambaram
ML Engineer | LLM Systems, Scalable Inference & GenAI Infrastructure | Computer Vision | vLLM Contributor | MS @ Stony Brook
- Role
- Open Source Contributor at Vllm
- Location
- Stony Brook, NY, US
- LinkedIn followers
- 500 followers
About Thillai Chithambaram
ML Engineer building high-performance LLM systems and scalable generative AI infrastructure, focused on turning cutting-edge research into production-ready systems that operate at scale. My work sits at the intersection of model intelligence and systems engineering. I design agentic LLM pipelines, optimize inference at the kernel and serving level, and build end-to-end AI systems that prioritize latency, throughput, and reliability. From multi-agent reasoning systems to multimodal perception pipelines, I focus on pushing both capability and efficiency. What I do best: Building and optimizing LLM systems (RAG, agents, tool use), accelerating inference with KV cache optimization, quantization, and TensorRT, and deploying scalable architectures using vLLM, PyTorch, and distributed training (FSDP, DDP). Real-world impact: Delivered production-grade LLM systems that improved retrieval accuracy by 38% and reduced latency by 45%, built aerospace vision systems with 98.8% precision, and developed medical AI models ranking among top global benchmarks. Open source & systems mindset:Active contributor to LLM inference ecosystems including vLLM and llm-d, with contributions spanning scalable serving, performance optimization, and knowledge editing (EasyEdit) for improving factual control and model reliability. Core focus areas: LLM systems • Agentic AI • Scalable inference • Multimodal models • Distributed ML systems
Experience
Open Source Contributor
Feb 2026 — Present
Enhanced performance metrics for quantized LLMs by expanding support from 3 to 22+ quantization methods (including GPTQ, AWQ, BitsAndBytes, and FP8 variants). Refactored quantization parsing logic and introduced unified weight mappings to improve accuracy of FLOPs and memory estimation, with added test coverage for robustness.
Education
Vellore Institute of Technology
Bachelor of Technology - BTech, Computer Science and Engineering
2020 — 2024
Stony Brook University
Master of Science - MS, Data Science
Find verified contacts for anyone on LinkedIn
Unifers gives sales teams verified emails and direct dials, enriched profiles, and outreach that lands in the inbox.
Free plan included · No credit card required
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.