ML Engineer, LLM Training (evals)

Eka.Care; Bengaluru; On-site; INR 800,000 - 1,200,000; Full time

Benefits offered by this job

Medical Insurance & Accidental保险

Job summary

EkaCare in Bengaluru is seeking a motivated engineer to rotate across data, evals and training tooling, with direct mentorship from the team running large runs and growth into a pretraining or post-training specialization within the first year.

You will run evaluation harnesses, build data-quality tooling, monitor training runs, and create internal tools that speed up researchers, all while learning through paper clubs and practical responsibilities.

Qualifications

• 1–3 years (or even fresh graduate) with excellent coding; Python you're proud of, and the instincts to debug anything.
• ML fundamentals you can defend on a whiteboard, not just import.
• Evidence you build things nobody asked you to: repos, competition results, a tool your classmates used.
• Hunger. This team moves fast, and the learning curve is the point.

Responsibilities

• Run and extend the evaluation harness: benchmark sweeps, dashboards, regression tracking across checkpoints.
• Build data-quality tooling: filters, dedup checks, samplers, look at the data utilities.
• Monitor training runs and make their logs and metrics legible.
• Build internal tools that make researchers measurably faster.
• Learn relentlessly, paper clubs, ablation reviews, and real responsibility early.

Skills

Python

Debugging

ML fundamentals

Open source

Tools

Git

Job description

About EkaCare and the mission

EkaCare is India's connected healthcare platform: an EMR that doctors run their practices on, a personal health record used by millions of Indians, and one of the deepest integrations with India's ABDM digital-health rails. Our Parrotlet family of medical models already serves Indian doctors in production, and we open-source our work where it counts.

The role

You’ll rotate across data, evals and training tooling with direct mentorship from the people running the big runs and grow into a pretraining or post-training specialisation within your first year.

What you’ll do

• Run and extend the evaluation harness: benchmark sweeps, dashboards, regression tracking across checkpoints.


• Build data-quality tooling: filters, dedup checks, samplers, \"look at the data\" utilities everyone thanks you for.


• Monitor training runs and make their logs and metrics legible.


• Build internal tools that make researchers measurably faster.


• Learn relentlessly, paper clubs, ablation reviews, and real responsibility early.

What we look for

• 1–3 years (or even fresh graduate) with excellent coding; Python you're proud of, and the instincts to debug anything.


• ML fundamentals you can defend on a whiteboard, not just import.


• Evidence you build things nobody asked you to: repos, competition results, a tool your classmates used.


• Hunger. This team moves fast, and the learning curve is the point.

Bonus

• Open-source PRs; Kaggle or competitive programming.


• Coursework or projects in NLP/LLMs; interest in Indic languages.

Why this is a rare gig

• Open source, with your name on it : weights and technical reports ship publicly.


• India-scale mission : models for a billion people in their own languages.


• Compute that’s rare to fine : dedicated multi-node H200 training under a national grant.


• Small senior team: you work with the people who own the recipe.


• A live deployment path : Government institutes, EkaCare’s doctors and patients use what you ship.

Full-Time Employee Benefits

• Medical Insurance & Accidental Insurance