Open to opportunities

Marvin Oduor

@marvinoduor

Message

AI Operations & Prompt Engineer building safe, accurate LLM workflows and policy-aligned systems.

Kenya

Message

What I'm looking for

I’m looking to join an AI team where I can design policy-aligned system instructions, run LLM evaluation and edge-case testing, and ship reliable JSON/API integrations—partnering closely with safety and engineering to reduce non-compliant outputs.

I’m an AI Operations & Prompt Engineering specialist with 4+ years building, evaluating, and enforcing safe and accurate LLM-powered workflows. I translate ambiguous compliance edge cases into clear, actionable system instructions, and I’ve helped reduce non-compliant model outputs through rigorous testing and policy documentation.

Most recently, I contracted with multiple early-stage AI product teams to design and quality-assure LLM system instructions and API integration workflows for Slack and PayPal, improving measured response accuracy by 22% across multi-locale deployments. Before that, at Lionbridge AI, I drafted and maintained operational content policies across 10+ global locales, evaluated high-volume ambiguous compliance cases, and ran QA testing that improved classifier precision.

Experience

Work history, roles, and key accomplishments

Current

AI Operations & Prompt Engineer

Current

Independent Contractor

Sep 2023 - Present (2 years 9 months)

Designed and quality-assured LLM system instructions and API integration workflows for Slack and PayPal SaaS integrations, improving measured response accuracy by 22% across multi-locale deployments. Built and validated prompt templates, analyzed API/JSON schemas, and reduced policy-violating outputs through rigorous testing and documented content-handling guidelines.

LLM System Instruction Authoring LLM Evaluation PayPal Sandbox Integration Testing Content Policy & Trust Safety Filtering