Aravind Sundaresan
Founder & LLM Inference / ML Infrastructure Engineer · Anavind Labs (Independent Research Practice)
Location: Hyderabad, India · Operating independently since November 2023
Systems Engineering & LLM Infrastructure
With an engineering background rooted in large-scale distributed systems at Microsoft and Amazon, I founded Anavind Labs to investigate, benchmark, and harden the core LLM serving and GPU infrastructure stack.
Having designed systems ranging from multi-service enterprise governance platforms to real-time hardware test schedulers, I bring the same operational rigor to AI runtimes: declare every hardware baseline, distinguish empirical telemetry from analytical models, and treat negative results as primary systems insights.
Origin & Philosophy: Named after Anantha and Aravind. In Sanskrit, Ananta (unbounded scale) meets Vinda (discovery and attainment)—reflecting our core focus on making unbounded compute attainable through deterministic systems engineering.
Every scheduler, runtime patch, kernel optimization, and benchmark published under Anavind Labs is conceived, implemented, profiled, and open-sourced for peer verification.
Research Manifesto
1. Precision Over Hype
No marketing superlatives. Every performance claim explicitly details the model checkpoint, concurrency profile, batching policy, input/output sequence distributions, and exact GPU topology.
2. Explicit Separation of Fact from Model
Measurements derived from discrete-event simulations, analytical projections, or synthetic trace replays are clearly designated as modeled data. Live hardware telemetry stands on its own.
3. Disconfirmed Results Carry Equal Value
Negative results are primary systems insights. When an intricate scheduler underperforms a simple FIFO queue, or an experimental kernel adds more launch overhead than wire savings, the disconfirmed result is published openly with complete profiling traces.
Let's Connect
I am currently open to conversations regarding senior/staff engineering roles, research collaborations, and production consulting in LLM inference serving, disaggregated architectures, and GPU cluster infrastructure.