Member of Technical Staff - Post-training (Data Infrastructure)

Role Overview

As a Member of Technical Staff in Post-Training Data Infrastructure, you will build and own the pipelines that produce the post-training data for our large language models (LLMs), from SFT and RL data to synthetic and human-annotated data. You will work closely with researchers and annotation teams so they can iterate on data as quickly as they iterate on models.

Key Responsibilities

- Build and maintain data pipelines for SFT, preference, and RL post-training data.

- Own dataset quality, versioning, and reproducibility.

- Turn research data recipes into production-grade, reusable components.

- Scale model inference for data generation and model-based grading.

- Design data formats and schemas that keep post-training datasets consistent and efficient to train on.

- Collaborate with researchers and engineers to make sure data pipelines keep pace with the needs of cutting-edge language models.

Qualifications

- Strong software engineering skills.

- Experience designing and building large-scale data pipelines for machine learning, ideally for LLM post-training.

- Track record of running distributed data processing and model inference at scale.

- Good understanding of post-training methods (e.g., SFT, RLHF, DPO) and how data shapes model behaviour.

- Excellent communication skills.

This is neither an exhaustive nor necessary set of attributes. Even if none of these apply to you, but you believe you will contribute to Cohere, please reach out. We value diverse backgrounds and perspectives at Cohere