Description

You will develop datasets and agentic workflows to improve Large Language Model performance on software engineering tasks.

This role is remote.

Responsibilities

• Deliver end-to-end agent use cases, including coding copilots and automation agents.
• Evaluate model-generated code responses by reviewing diffs for correctness, quality, and efficiency.
• Identify edge cases and ambiguities in model behavior through collaborative testing.
• Provide structured, detailed written rationales for code ranking decisions.

Required Skills

• 5+ years of software engineering experience.
• 2+ years of continuous experience at a top-tier product company (Google, Stripe, Amazon, Apple, Meta, Netflix, Microsoft).
• Expertise in building and deploying scalable, production-grade full-stack applications.
• Deep understanding of software architecture, design, and debugging.
• Proven ability to perform rigorous code reviews and assess maintainability.
• Strong oral and written communication skills for technical evaluation.
• Ability to commit to a minimum of 10 hours per week with partial PST overlap.

Preferred Skills

• Experience with LLMs and agentic workflows.

Key Skills

Education

Not specified