[Cornelius Kibet Kiptoo]
AI Training Specialist | Data Annotation Expert |Freelance
Professional Summary
Results-driven freelance AI training specialist with 4+ years of experience annotating,
labeling, and curating high-quality datasets for machine learning models. Proven track
record delivering 50+ projects for computer vision, NLP, audio transcription, and
multimodal AI systems on platforms Appen, Remotasks,QA World Analyst and Upwork.
Seeking long-term freelance or full-time opportunities in AI data preparation and model
evaluation.
Core Skills
- Image & Video Annotation (bounding boxes, segmentation, keypoints, 3D cuboids)
- Text Annotation (NER, sentiment, intent classification, summarization)
- Audio Transcription & Labeling
- LLM Response Evaluation
- Quality Assurance & Data Validation
- Tools: CVAT, Labelbox, Dataloop
- Languages: English ,Kiswahili [Native], Spanish[Basic]
Freelance Experience
Freelance AI Training & Data Annotation Specialist
Self-employed (via Upwork, Appen, Remotasks,QA WorldAnalyst)
2019 – Present
- Completed 50+ AI training projects
Image/Video projects:
– Annotated 2,000+ images for autonomous driving (LiDAR + camera fusion)
– Performed semantic segmentation
– Labeled retail product detection
NLP & LLM projects:
– Rated and rewrote 5,000+ LLM responses for helpfulness, safety, and style
– Built golden datasets for intent recognition and chatbot training
– Performed fact-checking
Audio projects:
– Transcribed and diarized 40+ hours of multilingual speech
– Labeled emotion, accent, and background noise for voice assistants
Notable Clients & Platforms (non-NDA)
- Appen – Multilingual speech and search relevance
- Remotasks – LiDAR annotation and 3D point-cloud labeling
- Various projects on Upwork
Education
Bachelor of Science in Computer Science
[Strathmore University]
Graduated: [2018]
Relevant coursework: Machine Learning, Data Structures
Additional Information
- Availability: 30–40 hours/week
- Rate: $18–45 USD/hour depending on task complexity
- Portfolio available upon request (screenshots of dashboards, accuracy metrics,
sample annotations – respecting NDAs)
- References available from long-term clients