Open to opportunities

Oluwaferanmi Ekundayo

@oluwaferanmiekundayo

Message

SFT & RLHF specialist aligning LLMs safely at scale.

Zimbabwe

Message

What I'm looking for

I’m looking for a role where I can own SFT + RLHF training and model evaluation—building high-quality rubrics, improving annotation workflows, and strengthening alignment, safety, and factuality through structured, human-in-the-loop feedback.

I’m an AI training specialist with over two years of hands-on experience across the full model development pipeline—from supervised fine-tuning (SFT) demonstration data creation through RLHF workflow execution—shaping large language model behaviour at scale. I focus on crafting high-quality SFT training examples, preference rankings, reward signals, and applying structured evaluation rubrics within production-scale pipelines.

I bring deep proficiency in hallucination detection, failure mode identification, and bias assessment across diverse prompt distributions. I design and apply evaluation frameworks for factuality, instruction-following, reasoning depth, coherence, and safety, maintaining high consistency across thousands of annotations. I also deliver granular feedback that improves model truthfulness and alignment.

My edge comes from combining an engineering foundation in systematic root-cause analysis with a psychology-informed understanding of human intent. Through annotation workflow optimisation, I improve throughput, inter-annotator agreement, and scalability in human-in-the-loop systems. I enjoy collaborating remotely and across functions to produce clear technical feedback and evaluation reporting that drives continuous model improvement.

Experience

Work history, roles, and key accomplishments

Current

Associate Consultant

Current

Evo & Hogg Limited

Apr 2025 - Present (1 year 2 months)

Diagnosed operational inefficiencies using structured analytical frameworks and delivered data-driven improvement strategies. Produced detailed technical reports for senior stakeholders and collaborated across disciplines under time pressure.

Root Cause Analysis Process Improvement Data Strategy Technical Reporting Cross Functional Collaboration

Current

AI Training Specialist

Current

Multiple Platforms

Aug 2023 - Present (2 years 10 months)

Produced high-quality supervised fine-tuning (SFT) prompt-response demonstration data and delivered human feedback (preference rankings) to inform RLHF reward model training and iterative alignment. Designed and applied evaluation rubrics for quality, factuality/hallucination detection, bias, and safety, maintaining consistency across thousands of annotations.

Evaluation Rubric Design Hallucination Detection Annotation Workflow Optimisation

Maintenance Operative

Brioche Pasquier UK

Mar 2023 - Mar 2025 (2 years)

Led teams in high-precision, high-throughput production environments, maintaining rigorous quality and accuracy standards. Applied fault diagnosis and root-cause analysis and implemented process optimisation to improve consistency and output reliability.

Team Leadership Quality Control Root Cause Analysis Process Optimisation Reliability

Junior Process Engineer

Ken Lewis Engineering

Sep 2021 - Aug 2022 (11 months)

Applied data-driven process optimisation and scientific methodologies to improve system performance while maintaining strict compliance with quality and operational standards. Supported structured, evidence-based problem solving in controlled manufacturing environments.

Process Optimisation Data Driven Systems Performance Quality & Compliance Manufacturing Operations Root Cause Analysis

Education

Degrees, certifications, and relevant coursework

University of Buckingham

Bachelor of Science, Psychology with English Literature

Earned a BSc in Psychology with English Literature, developing insight into human cognition, language interpretation, and behavioural intent to support nuanced, human-aligned evaluation and preference judgments.

University of South Wales

Bachelor of Engineering, Mechanical Engineering

Completed a BEng in Mechanical Engineering, strengthening systematic thinking, structured problem-solving, and root-cause analysis that underpin later model evaluation and failure mode analysis work.