I’m a machine learning engineer focused on applied ML and high‑performance LLM inference & serving.
Lately I’ve been helping shape Vajra, building Veeksha for evaluation, and advising startups on applied AI. Below is the short, human version of my timeline.
Machine Learning Systems Engineer at Vajra, an upcoming open‑source inference engine. Designed and worked on quantization & MoE support, experiment infra, CI, and telemetry. Lead developer of Veeksha (LLM performance & quality evaluation).
Advising companies on end‑to‑end applied AI: setting requirements, data, training, eval and serving.
While doing my MSc, I was a visiting researcher at the Barcelona Supercomputing Center. I briefly studied advanced batching for LLM inference.
Designed and shipped ML systems for asset & credit risk. Productionized work with measured impact in the tens of millions.
For ~2 years, I was interned and employed in various ways by the two institutions. Participated in two distinct projects.
First cohort of Spain’s data science program. Thesis and internships centered on ML.
Me and Cookie 🍪