We're looking for strong LLM research engineering candidates to join our startup. We've built a super verticalized retrieval augmented generation (RAG) pipeline and are pushing the limits of what's possible here and co-discovering new limitations/features with our customers every day.
You will work directly with the founding team on continously improve the core RAG pipeline, iterate on the evalution suite and implement state-of-the-art techniques (e.g., fine-tuning cross-encoders). In addition to the founding team, you'll have support from a number of leading academics in the field that are all close advisors (incl. Douwe Kiela, author of the original RAG paper).
Here are a few signs you'd be a good fit:
- You've built RAG pipelines before
- You really care about LLM evals
- You're a researcher (PhD grad) that want to start applying the knowledge in a fast-growing start-up setting