Looking for a job

Chris Ku

@chrisku

Message

Senior machine learning engineer building efficient computer vision and on-device generative AI.

Zimbabwe

Message

What I'm looking for

I’m looking to lead production ML that ships—optimizing computer vision and generative AI for latency, cost, and on-device performance, while collaborating with product teams to translate customer needs into measurable outcomes.

I’m a Senior Machine Learning Engineer with 10+ years in computer vision, deep learning, and production ML systems, and I enjoy turning research into shipped, measurable impact. At Raspberry AI, I developed and deployed production generative AI for fashion design automation, including a conversational design agent built with MCP and Google Vertex AI, plus brand-specific LoRA fine-tuning for Fortune 500 apparel and footwear.

I also specialize in efficient model optimization—quantization-aware training, distillation, pruning, and on-device deployment—backed by my work co-developing MobileNetV4 and shipping vision models through TensorFlow Model Garden at Google. I’ve built end-to-end pipelines for high-quality training data, semantic asset retrieval for sub-second similarity search, and inference optimizations that reduced infrastructure costs by 40%, while using ML evaluation frameworks and A/B testing to validate visual fidelity and customer satisfaction.

Experience

Work history, roles, and key accomplishments

Current

Senior Machine Learning Engineer

Current

Raspberry AI

Feb 2024 - Present (2 years 4 months)

Developed and deployed production generative AI models for fashion design automation, converting sketches into photorealistic product renderings. Built a conversational design agent with MCP/Vertex AI and LoRA personalization, including a Fashion-CLIP semantic retrieval system and inference optimizations that reduced infrastructure costs by 40% and cut sample development time by 60%.

MCP Vertex AI Fashion CLIP Embeddings Training Data Pipelines (EXIF Color Normalization)Inference Optimization (Batching Caching)ML Evaluation And A B Testing

Senior ML Software Engineer

Google

Jun 2020 - Dec 2023 (3 years 6 months)

Co-developed MobileNetV4, achieving 87% ImageNet accuracy at 3.8ms inference latency on Pixel 8 EdgeTPU. Served as a core developer for TensorFlow Model Garden (used by 1B+ users) and implemented quantization-aware training for INT8 deployment with 3x latency reduction.

MobileNetV4 Tensorflow Lite INT8 Deployment Anomaly Detection

Senior Data Engineer

Valtech

Aug 2017 - May 2020 (2 years 9 months)

Designed and maintained large-scale ETL pipelines in Python, Spark, and Airflow, transforming multi-terabyte datasets into AWS Redshift and Snowflake. Improved query performance and reduced reporting latency by 40%, while building CI/CD and infrastructure automation with GitLab and Terraform for staging/production deployments.

Python Apache Spark Apache Airflow AWS RedShift Snowflake ETL CI CD Gitlab Terraform

Research Scientist

eBay

Jun 2014 - Jun 2017 (3 years)

Architected and deployed large-scale visual search systems for 1B+ live product listings using deep learning for image classification and similarity matching. Created ModaNet (55,176 polygon-annotated images across 13 categories) and built real-time indexing and hashing for retrieval at scale, alongside object detection and semantic segmentation advances.

Visual Search Deep Learning Real Time Indexing And Hashing Object Detection (YOLO Faster R CNN SSD)Dataset Creation (ModaNet)Visual Saliency Detection Image Retrieval At Scale