Open to opportunities

Toheeb Sikiru

@toheebsikiru

Message

AI QA Engineer specializing in AI safety and LLM testing with 3+ years and 500+ bug reports.

Nigeria

Message

What I'm looking for

I’m looking to join a team that ships AI responsibly—where I can run rigorous AI safety and LLM testing, build strong prompt test coverage, and improve quality through clear, reproducible bug reports across products.

I’m a Remote QA Engineer focused on AI safety and LLM testing, with 3+ years of experience and 500+ accepted bug reports on Test IO. I’m known for catching critical guardrail failures—especially self-harm detection gaps and broken crisis flows—in live EdTech and mental health products.

I deliver fully documented, high-quality defect reports with zero supervision, using structured prompt test sets and a multi-criterion scoring framework to quantify AI behavior. My work includes functional regression, exploratory testing, and cross-browser/cross-platform manual QA across web, iOS, and macOS, with attention to usability and accessibility (WCAG 2.1 AA).

I also bring hands-on quality expertise from AI training data projects, where I’ve led annotation consistency enforcement and video triage to improve dataset reliability. Across distributed teams, I manage project review workflows remotely while maintaining strong quality standards from intake to final validation.

Experience

Work history, roles, and key accomplishments

Current

QA Team Lead, Data Collection

Current

Applause App Quality

Jun 2023 - Present (3 years)

Led quality review for AI training data submissions by triaging participant-collected videos and enforcing annotation quality standards for AI model training pipelines. Managed remote team output and project review workflows, maintaining annotation consistency across large-scale data collection to improve dataset reliability.

Video Triage Annotation Consistency Enforcement Data Quality Team Leadership Remote QA Workflow Management

Current

Senior Software Tester

Current

TestIO Inc.

Jul 2022 - Present (3 years 11 months)

Delivered 500+ accepted bugs across 6 platforms, including 4 critical AI safety guardrail failures (self-harm detection gaps and broken crisis pathways) in K-12 EdTech and child-facing products. Built structured prompt test sets and used a five-criterion scoring framework to quantify AI behavior, driving cross-browser and Arabic/RTL localization regressions.

Adversarial And Benign Prompting Response Validation With Scoring Frameworks Test IO Bug Reporting