Plsn The Research Engineer, ML Infrastructure builds and owns distributed training systems across GPU clusters to support AI model development. This role focuses on reliability, performance profiling, and tooling to enable reproducible experiments.
Responsibilities
Build and own distributed training systems across GPU clusters. Job launchers, checkpointing, recovery and monitoring should keep experiments moving when something fails.
Profile training end to end. Find bottlenecks in data loading, communication, memory and compute, then fix the ones that matter.
Build the tooling to launch, track and compare experiments. Make runs reproducible and shorten the path from an idea to a result.
Build reliable data pipelines for training and evaluation. Own data quality, throughput and the ability to reproduce a run.
Debug failures across GPUs, networking, numerics and data. Build systems that recover cleanly instead of leaving researchers to start over.
Implement and tune distributed parallelism where it helps. Understand the hardware and the tradeoffs rather than treating the training stack as a black box.
Work directly with research to build infrastructure before it becomes a bottleneck. You own the systems from the cluster to the training loop.
We are a small team with a high bar and very little process. You get a real problem and the trust to go after it, and you decide how it gets solved. We think carefully and move fast.
When you build something here, you own it. The system, the results, and what happens when it breaks. Nobody else is going to pick it up for you.
Qualification
A Research EngineerA builder with public work to showHigh agencyTapped inYou care how and why models workFast to learnYou thrive in small
Required
A Research Engineer, ML Infrastructure who thinks from first principles and builds things that work outside the lab. You care about why people want something and why models behave the way they do, and you care that what we ship works when it actually matters. If you have built products, shipped software, talked to customers or run outreach, and you are drawn to the hardest reliability problems in AI, this is the role.
This is a role for someone early in their career who has already built real things. We care more about how fast you have been learning than where you have been.
A builder with public work to show. Repos, hardware, a product, a paper, a competition result, a project or company you ran at school. We will read it, so make it good.
High agency. You want to build a company from the ground up. You start things without being asked, you finish them, and you take 100% ownership of the systems you build.
Tapped in. You know what is happening in the tech ecosystem around you and around the world. You spend time on X, you follow what people are building, and you know and connect with the people doing it.
Comfortable working from first principles. You can define a problem, talk to the people who have it, and build the minimal system that proves what is possible.
You care how and why models work, and you care that they actually work when the stakes are real.
Fast to learn. Your last year looks very different from the year before it.
Solid in Python and at home with ML tooling. You do not have to be a researcher.
You thrive in small, focused teams that value autonomy, speed and tight collaboration over process.
You think AI for critical systems is one of the most consequential problems in the field right now and you want to be early to solving it.
Students and recent graduates are welcome. We do not require a degree or a set number of years of experience.
