Remote LLM Agentic Trainer - Programmer

About the role

We are looking for a remote LLM Agentic Trainer with strong programming skills to validate and improve large‑scale training datasets used in agentic AI systems. The role combines data quality assurance with the creation of JSON‑based function‑calling tasks, supporting cutting‑edge AI research and product development.

Key responsibilities

• Validate large‑scale training datasets and enforce strict quality standards.
• Create, manipulate, and structure JSON tasks for function‑calling workflows.
• Collaborate with AI researchers to align datasets with agentic AI use cases.
• Monitor data accuracy, consistency, and integrity across multiple projects.
• Document best practices and contribute to dataset preparation guidelines.

Required profile

• Strong proficiency in at least one programming language (e.g., Python, JavaScript, Java).
• Solid understanding of SQL for querying and managing datasets.
• Experience handling JSON data structures.
• Exceptional attention to detail and focus on data quality.
• Interest or exposure to agentic AI, large language models, or NLP.

Required skills

• Python
• JavaScript
• Java
• SQL
• JSON

What we offer

• Fully remote work environment.
• Flexible weekly commitment (20, 30 or 40 hours) with overlap in PST.
• Opportunity to work on cutting‑edge AI projects with leading LLM companies.
• Contract‑based engagement with short‑term extensions possible.