For employers

Hire this AI Trainer

Sign in or create an account to invite AI Trainers to your job.

LoginCreate account
Invite to Job
Fisayo Adedoyin

Fisayo Adedoyin

MLLM, LLM Evaluation, Text Generation (image and video)

NIGERIA flag
Lagos, Nigeria
$40.00/hrExpertCVATOtherInternal Proprietary Tooling

Key Skills

Software

CVATCVAT
Other
Internal/Proprietary Tooling

Top Subject Matter

No subject matter listed

Top Data Types

ImageImage
TextText
VideoVideo

Top Label Types

Question Answering
Red Teaming
RLHF
Segmentation
Text Generation

Freelancer Overview

Generative AI Team Lead & RLHF Specialist with 4+ years of experience refining foundational models, including high-stakes SFT/RLHF work for Nvidia. Expert in Model Humanization (Tone, Style, Natural Speech) and Multi-Modal Data Ops, consistently maintaining 95%+ accuracy benchmarks for Meta AI. Proven leader in AI Governance and Red Teaming, managing teams of 100+ to ensure models are ethically aligned, logically consistent, and safety-tested.

ExpertEnglish

Labeling Experience

Team Lead, Generative AI (MMLLM & LLM Operations)

Internal Proprietary ToolingTextClassificationQuestion Answering
Led a team of 100+ annotators and 24 auditors in the development of training programs for Meta AI's Multi-Modal models. Designed logic rubrics and audit scoreboards that increased overall annotation accuracy to a 95% benchmark while ensuring ethical handling of sensitive datasets.

Led a team of 100+ annotators and 24 auditors in the development of training programs for Meta AI's Multi-Modal models. Designed logic rubrics and audit scoreboards that increased overall annotation accuracy to a 95% benchmark while ensuring ethical handling of sensitive datasets.

2024

Python Coding Specialist & AI Trainer

Internal Proprietary ToolingComputer Code ProgrammingComputer Programming Coding
In a short term, I personally verified the logical accuracy of model-generated code, assessed code quality and clarity, and captured reproducible error traces to document failure modes.

In a short term, I personally verified the logical accuracy of model-generated code, assessed code quality and clarity, and captured reproducible error traces to document failure modes.

2025 - 2025

RLHF & SFT Specialist (Nvidia Project)

Internal Proprietary ToolingTextText GenerationRLHF
Performed Reinforcement Learning from Human Feedback (RLHF) and Supervised Fine-Tuning (SFT) to refine model parameters for accuracy and safety. Systematically ranked thousands of model outputs based on "Triple H" criteria (Honesty, Helpfulness, Harmlessness) and rewrote prompt answers to meet strict human-like interaction rubrics.

Performed Reinforcement Learning from Human Feedback (RLHF) and Supervised Fine-Tuning (SFT) to refine model parameters for accuracy and safety. Systematically ranked thousands of model outputs based on "Triple H" criteria (Honesty, Helpfulness, Harmlessness) and rewrote prompt answers to meet strict human-like interaction rubrics.

2024 - 2024

Quality Assurance & Training Specialist (AI Safety)

Internal Proprietary ToolingTextEmotion RecognitionRed Teaming
Conducted safety-critical Red Teaming and adversarial testing to identify and mitigate bias within Large Language Models. Reduced operational risk to a 5% error rate by embedding governance checkpoints within QA audits and leading SME training sessions on ethical annotation standards.

Conducted safety-critical Red Teaming and adversarial testing to identify and mitigate bias within Large Language Models. Reduced operational risk to a 5% error rate by embedding governance checkpoints within QA audits and leading SME training sessions on ethical annotation standards.

2023 - 2024

Video Annotation & SRT Specialist

Internal Proprietary ToolingVideoAction RecognitionTracking
Worked hands-on with SRT software to perform precise temporal labeling and time-synced data tagging for Meta AI video datasets. Used CVAT for frame-by-frame object tracking and action recognition, while simultaneously leading a team of 100+ to maintain consistent labeling standards across high-volume video streams.

Worked hands-on with SRT software to perform precise temporal labeling and time-synced data tagging for Meta AI video datasets. Used CVAT for frame-by-frame object tracking and action recognition, while simultaneously leading a team of 100+ to maintain consistent labeling standards across high-volume video streams.

2021 - 2023

Education

O

Obafemi Awolowo University

Bachelor of Science, Demography and Social Statistics

Bachelor of Science
2015 - 2019

Work History

No Work History added yet

Fisayo A. hasn’t added any Work History to their OpenTrain profile yet.