Ai Research Engineer • Mid-Level

Who you are

You have hands-on ML experience, particularly in training, finetuning, or evaluating models — ideally with a focus on embeddings or large language models (LLMs). Your strong engineering fundamentals allow you to build reliable systems, and you are proficient in Python and Rust, as well as working with distributed pipelines and GPU/cluster jobs.

You enjoy diving into data, building evaluation sets, and inspecting edge cases — your creativity shines through in designing measurement strategies that effectively assess search engine performance. You are excited about the challenge of defining what 'good search' means in the context of AI and are eager to contribute to shaping the future of search technology.

What you'll do

In this role, you will design and build the evaluation stack at Exa, focusing on how to evaluate search engines in an LLM world. You will create comprehensive and effective evaluation suites that reflect real-world usage by developers, agents, and humans. Your work will involve collaborating closely with ML researchers, data engineers, and infrastructure engineers to establish feedback loops that enhance search models.

You will write a manifesto detailing what perfect search means and design benchmarks, agentic tasks, and real-world test suites. Your contributions will directly influence the optimization strategies we adopt for our search engine, ensuring that we can filter the world's knowledge effectively and efficiently.

What we offer

This is an in-person opportunity based in San Francisco, and we are open to sponsoring international candidates. You will be part of a team that is at the forefront of AI technology, working on one of the most ambitious indexing operations in the world. We offer a collaborative environment where your ideas and contributions will be valued as we strive to build the search engine for the age of AI.