Engineering Notes

Hi — I’m Pranav. I write about production AI infrastructure, LLM benchmarks on real-world workloads, and observability engineering.

I currently build AI-SRE pipelines using OpenClaw, NVIDIA NIM, and a fleet of free-tier models. Previously I shipped Elixir systems at scale for a fintech.

Find me on GitHub · LinkedIn.

Posts

• Apr 28, 2026

  I benchmarked 21 NVIDIA NIM free-tier models on real production AI-SRE workloads. Here's the architecture that emerged.

subscribe via RSS