Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

I am a Senior Machine Learning Engineer and co-founder with 5 years of experience building and deploying production ML systems, specializing in real-time computer vision and ML infrastructure. I co-founded Videolity and secured £500K+ funding by leading technical due diligence and shipping high-performance prototypes. I own end-to-end data strategy and ML pipelines, from data collection/labeling to training, deployment on AWS/Docker, and rigorous evaluation. I have delivered LLM agent systems using RAG and tool calling, and integrated advanced visual recognition workflows for robust retail video analysis.…I am a Senior Machine Learning Engineer and co-founder with 5 years of experience building and deploying production ML systems, specializing in real-time computer vision and ML infrastructure. I co-founded Videolity and secured £500K+ funding by leading technical due diligence and shipping high-performance prototypes. I own end-to-end data strategy and ML pipelines, from data collection/labeling to training, deployment on AWS/Docker, and rigorous evaluation. I have delivered LLM agent systems using RAG and tool calling, and integrated advanced visual recognition workflows for robust retail video analysis.

AŞİR ŞİYAM ÇİÇEK

Data Scientist, AI Engineer, Full Stack Developer, +2





I am a Senior Machine Learning Engineer and co-founder with 5 years of experience building and deploying production ML systems, specializing in real-time computer vision and ML infrastructure. I co-founded Videolity and secured £500K+ funding by leading technical due diligence and shipping high-performance prototypes. I own end-to-end data strategy and ML pipelines, from data collection/labeling to training, deployment on AWS/Docker, and rigorous evaluation. I have delivered LLM agent systems using RAG and tool calling, and integrated advanced visual recognition workflows for robust retail video analysis.…I am a Senior Machine Learning Engineer and co-founder with 5 years of experience building and deploying production ML systems, specializing in real-time computer vision and ML infrastructure. I co-founded Videolity and secured £500K+ funding by leading technical due diligence and shipping high-performance prototypes. I own end-to-end data strategy and ML pipelines, from data collection/labeling to training, deployment on AWS/Docker, and rigorous evaluation. I have delivered LLM agent systems using RAG and tool calling, and integrated advanced visual recognition workflows for robust retail video analysis.

Available to hire

I am a Senior Machine Learning Engineer and co-founder with 5 years of experience building and deploying production ML systems, specializing in real-time computer vision and ML infrastructure.

I co-founded Videolity and secured £500K+ funding by leading technical due diligence and shipping high-performance prototypes. I own end-to-end data strategy and ML pipelines, from data collection/labeling to training, deployment on AWS/Docker, and rigorous evaluation. I have delivered LLM agent systems using RAG and tool calling, and integrated advanced visual recognition workflows for robust retail video analysis.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Language

Turkish

Fluent

English

Fluent

French

Beginner

Work Experience

Co-founder & Senior Machine Learning Engineer at Videolity

April 1, 2022 - December 1, 2025

Co-founded and architected Videolity’s end-to-end ML platform on AWS (EC2, SageMaker, ECR) and Docker, delivering production-grade, real-time retail computer vision from data ingestion to deployment. Secured £500K+ funding (Techstars, Forward VC) by leading technical due diligence and shipping high-performance prototypes. Designed and deployed an LLM-powered customer support agent using OpenAI, Hugging Face, and LangChain with tool calling, RAG, and automated evaluation pipelines to reduce hallucinations and policy violations. Built production pipelines (CNNs, Transformers) and integrated YOLO + DeepSORT + DINOv2 for clothing detection/tracking in fast-cut footage, boosting mAP/IDF1 from ~0.60 to ~0.92. Developed a cross-domain visual verification pipeline (SAM2, DINOv2/CLIP, LoFTR) achieving ~95% accuracy on a human-verified internal dataset, robust to lighting/viewpoint changes. Fine-tuned DINOv2/CLIP on domain data for retrieval/verification. Owned data strategy end-to-end: data

Machine Learning Contractor at Shade and Rose Ltd

June 1, 2020 - April 1, 2022

Integrated and deployed ML models into production-grade systems for multiple startups, collaborating with distributed engineering teams to ensure reliability, performance, and maintainability. Delivered custom ML solutions across agriculture, fitness, and mobile applications. Improved drone-based crop-quality detection using optimized computer-vision algorithms. Built a fitness mobile application with integrated pose-detection AI, deployed via an AWS-backed production stack.