~/about

I’m Pranay. I build distributed systems, then break them on purpose to find out where the tail latency hides.

The short version: electronics undergrad in Bengaluru → robotics research at ARTPARK @ IISc (log aggregation and safety systems for warehouse robots) → two years at MakeMyTrip running A/B infrastructure at India-travel scale, where I learned that a slow p99 is a personal insult (the CV has what I did about it). Then an MS in Computer Science at UC Santa Cruz, finished in August 2026, where my research became faults-lab: a platform for injecting faults at service boundaries and isolating each service’s contribution to tail latency under load. I’m still building it.

What I care about, in order: correctness, p99s, and APIs that don’t make the next engineer’s life worse. I write Go and Rust by preference, Java when the legacy platform demands it, and I instrument everything — if it doesn’t emit traces, it didn’t happen.

on naming things

Every system I build gets a name from classical mythology, and the names are load-bearing:

• atropos — the Fate who cuts the thread. It’s the fault injector; it decides when your request dies.
• zeus — throws the lightning. Load orchestration; it commands the Vegeta attacks.
• manteion — the oracle at Delphi. The experiment brain that observes everything and answers questions.

The chat widget in the nav is called the oracle for the same reason. I am aware this is a disease. I have no plans to treat it.

elsewhere

When not staring at Grafana: I’m somewhere between San Francisco and Santa Cruz, and I built slug-mcp partly so my own AI tools could tell me the surf conditions and whether the campus gym is full. I also TA’d UCSC’s software engineering sequence (CSE115A/B/C), mentoring teams on projects sponsored by Nutanix and Keysight.

If any of this sounds like your kind of problem, say hello.