LLM Trainer – Agent Function Calling
Location: Nigeria, Kenya, Egypt, Ghana
Job Type: Contractor
Contract Duration: 6 weeks
Commitment: 20, 30, or 40 hours per week
Schedule: Minimum 4 hours per day with 4 hours of PST overlap
About the Role
We’re looking for technically strong professionals to help improve the capabilities of advanced Large Language Models (LLMs), with a focus on AI agents and function calling.
You’ll create realistic, multi-turn conversations between users and AI assistants that use tools such as calendars, email, maps, and cloud storage to complete tasks.
The work involves simulating both sides of the interaction and creating high-quality examples that demonstrate effective reasoning, tool usage, contextual understanding, and natural communication.
What You’ll Do
Design realistic multi-turn conversations involving AI assistants and external tools.
Simulate both the user and assistant sides of conversations.
Create and evaluate appropriate function calls when tools are required.
Develop scenarios involving applications such as calendars, email, maps, and drive.
Demonstrate how an AI assistant should handle complex tasks and real-world decision-making.
Create examples where assistants successfully complete tasks, recognize limitations, or respond appropriately without using tools.
Ensure conversations follow required formatting, quality, and consistency standards.
Refine examples based on reviewer feedback.
Collaborate with reviewers and other contributors to maintain high-quality deliverables.
Requirements
Strong technical reasoning and logical problem-solving skills.
Ability to break complex tasks into clear, realistic steps.
Understanding of APIs, JSON, data formats, and tool-based workflows.
Experience with any programming language or technical stack.
Excellent written English and ability to create natural, coherent dialogue.
Strong creativity and attention to detail.
Ability to consistently follow detailed guidelines and formatting requirements.
3+ years of professional experience in a technical or analytical field.
Nice to Have
Experience working with LLMs or generative AI.
Familiarity with AI assistants or agentic systems.
Experience with function calling or tool-use frameworks.
Experience creating datasets, evaluations, or training examples for AI systems.
Engagement Details
Hours: 20, 30, or 40 hours per week.
Minimum: 4 hours per day.
Time Zone: 4 hours of overlap with PST required.
Contract: 6 weeks.
Engagement: Contractor assignment.
No medical or paid leave.
Who This Role Suits
This opportunity is a strong fit for technical professionals who understand how software, APIs, and structured data work and can translate that knowledge into realistic conversations between people and AI assistants.
You should be comfortable thinking through what an AI agent should do, which tools it should use, when it should use them, and how it should communicate with the user throughout the process.