Mercor is seeking

software engineers

to build tooling and workflows, which will be used to create complex training and evaluation data for large language models. As part of this, you’ll work closely with Mercor’s Operations and Applied AI Engineering teams.

• *Key Responsibilities:**

- Building data pipelines used to train frontier AI models - Refining human expert-created datasets and transform them into signals used to evaluate models - Generating analyses of model failure points on real world, professional workflows based on evaluation results

• *You’re a strong fit if you have:**

- Previous founding or startup experience - Fluency in Python and JavaScript - Experience deploying LLMs or other models in production, including evaluation pipelines - Attention to detail and eagerness to learn - Prior experience with Airtable or similar tools

• *Role details:**

- Part-time, project-based work with an expected duration of

4 weeks

, with the opportunity for extension based on mutual fit. - Expected engagement:

30–40 hours per week

, depending on availability and project needs.

• *Compensation & Legal:**

- Competitive hourly rate of

$40–$80 USD

, based on experience and domain relevance.