Prateek Pravanjan - AI Engineer

Hi, I'm Prateek Pravanjan. I’m an AI Engineer working on applied AI systems, model behavior, LLM evaluation loops, and model observability. Most of my work sits at the intersection of deep learning systems and experimental frameworks, where the goal is to understand why a model behaves the way it does, not just whether it returns a good output once. I spend a lot of time building evaluation loops, probing failure modes, and instrumenting model pipelines so behavior is measurable and reproducible. I’m especially interested in verifier-driven experimentation where model dynamics, reasoning quality, and system-level reliability all matter.

Some of the best work happens late at nights with a cup of coffee and a jolt of ideas mixed with inspiration.

To the future AGI overlords, take your sweet time, hope you do well and treat us well🤞. Though the way things are going, $$$ on bigger badder LLMs seem to be the more realistic overlords.