LLM Trainer - Agent Function call
Core
Simulate high-quality multi-turn conversations between users and AI assistants utilizing function-calling tools to generate proprietary data for fine-tuning and benchmarking foundational LLMs.
Role type
LLM Trainer (Agent Function Call)
Builds
High-quality proprietary dialogue datasets for fine-tuning and evaluation of Large Language Models.
Domain
Artificial Intelligence / Large Language Models / Data Annotation
Deliverable
production ML models
Required skills
Python, Java, JavaScript, API integration, JSON data formats, technical reasoning, logical thinking, English writing
Preferred skills
Experience with LLMs, virtual assistants, function-calling frameworks
Responsibilities
Design multi-turn conversations simulating real interactions with tool-based apps (calendar, email, maps, drive), emulate user and assistant roles including tool calls, select appropriate tool usage for logical flow, craft dialogues demonstrating natural language and contextual understanding, iterate on examples based on feedback, collaborate with peers to maintain consistency.
Seniority
Mid-level (3+ years experience)
