Chepkemoi R.

Key Skills

Software

CloudFactory

CrowdSource

Snorkel AI

Roboflow

Lionbridge

Humanatic

Clickworker

Anno-Mage

Appen

Argilla

Axiom AI

CVAT

Data Annotation Tech

Datature

Diffgram

Top Subject Matter

E-commerce

Finance

Research

Top Data Types

Image

Text

Document

Top Task Types

Bounding Box

Segmentation

Polygon

Point/Key Point

Cuboid

Text Generation

Question Answering

RLHF

Red Teaming

Evaluation/Rating

Computer Programming/Coding

Transcription

Function Calling

Data Collection

Prompt + Response Writing (SFT)

Freelancer Overview

1. Code Quality Evaluation & Pull Request Review — TechOps Open Source Community (2021–Present) Reviewed 100+ pull requests across 50+ open-source repositories, evaluating code for correctness, logic, style, and edge-case handling. Provided structured written rationales for every decision — directly mirroring the annotation feedback format used in RLHF pipelines. Contributed to 50+ repositories with a combined 1,000+ stars. 2. LLM Response Evaluation & Prompt Testing — Personal AI Training Practice (2024–Present) Evaluated and ranked AI-generated responses across coding, reasoning, and natural language tasks. Wrote targeted prompts to surface model weaknesses, scored outputs for helpfulness, accuracy, and safety, and documented findings in structured annotation format. Applied preference ranking across English and Swahili outputs. 3. Technical Content Annotation & Tutorial Production — YouTube / Medium / Dev.to (2023–Present) Produced step-by-step annotated instructional content across Java Spring Boot (15,000+ views) and React Hooks (10,000+ views). This work required precise labeling of concepts, identifying misconceptions, and structuring information so it could be used as clean training material — closely aligned with instruction-tuning data creation. 4. Rubric-Based Developer Assessment — Safaricom Women in Tech / West Africa WiT (2022–Present) Designed and ran coding workshops for 200+ participants, evaluating submissions against defined criteria and providing detailed corrective feedback. Mentored 25+ junior developers with structured, consistent scoring — equivalent to applying annotation guidelines at scale. 5. Multilingual Technical Documentation — Safaricom PLC (2023–Present) Created and reviewed technical documentation for systems serving 45M+ users across Kenya and East Africa, working across English and Swahili contexts. Ensuring accuracy, clarity, and consistency in multilingual technical content is a core skill for cross-lingual NLP annotation projects.

Education

Work History