Hi, I'm Truong Giang. I build machine learning that actually ships.
A London-based AI & ML consultant. For five years I've put real systems into production — defect detection at 98% precision, recommendation engines, RAG assistants — and I write about all of it as I go.
MSc Data Science (Distinction) · 5+ yrs shipping ML · GCP certified · London, UK
Ways I can help
End-to-end, from a half-formed idea to a system your team can actually run. Pick a slice or the whole thing.
LLM fine-tuning & RAG systems
Custom LLMs and retrieval-augmented generation with grounded, low-latency answers. Multi-agent setups, enterprise chatbots, document processing, knowledge management.
Computer vision development
Image and video analytics for manufacturing and retail. Defect detection at 98% precision in 0.2s, object detection, visual inspection, and quality-control automation.
MLOps & pipelines
Production ML infra on GCP and AWS. Automated training with Vertex AI and Airflow, CI/CD, deployment, monitoring, and continuous retraining that doesn't fall over.
Predictive analytics & recsys
Forecasting, segmentation, and recommendation engines — up to 92% precision — turned into decisions and measurable growth, not dashboards nobody opens.
AI strategy & consulting
Roadmaps tied to business goals. Opportunity assessment, feasibility, and implementation planning — ROI-focused, with clear milestones and honest trade-offs.
Healthcare & medical AI
Clinical AI with the compliance to match. GCP-certified, with research on Parkinson's treatment prediction — medical data analysis and decision support.
Things I've built
A few projects across manufacturing, retail, and healthcare — plus what I build for fun.
Multi-agent customer support
A working multi-agent support system on Gemini — a Policy Agent (RAG) and a tool-using Ticket Agent — built end to end, then rebuilt in LangGraph to compare.
- ›Master + Policy (RAG) + Ticket agents
- ›15 scenarios, LLM-as-a-judge eval
- ›Open source, 8-part write-up
Fashion recommendation system
Led 3 engineers to build a computer-vision recsys for fashion retail with RetinaNet and OpenCV, shipped across web and mobile.
- ›92% average precision in outfit detection
- ›Led a cross-functional team of 3
- ›Lifted click-rate and cross-sell
Semiconductor defect detection
A production defect-detection system for semiconductor QC, running on the factory floor in real time with a full ML workflow and WPF app.
- ›98% precision in defect identification
- ›0.2s inspection per image
- ›Real-time factory deployment
Parkinson's treatment research
Predictive models evaluating how well wearable and smartphone cues work in treating Parkinson's-related drooling — a contribution to clinical science.
- ›65% F1 in schedule prediction
- ›Deployed on AWS SageMaker
- ›Clinical data analysis & viz
A bit about me
I'm Truong Giang, the person behind twentytwotensors. I started it because I like the whole arc — sitting with a messy problem, finding the model that fits, and seeing it run in production where it has to keep working.
My work spans NLP, computer vision, and cloud-native ML infrastructure, delivered across manufacturing, retail, and healthcare. A semiconductor system at 98% precision in 0.2s an image; a fashion recommender at 92% precision where I led a team of three.
I bring the full loop — data engineering, model development, deployment, and the ongoing tuning afterwards — on GCP and AWS. And I write about it openly: the wins, the middling eval scores, all of it.
98% accuracy and 92% recommendation precision, in live systems.
London-based for on-site work, fluent in UK data-protection requirements.
From data engineering to monitoring — the whole ML lifecycle.
An active blog documenting real projects, honestly.
From the blog
Notes from real projects — what I'm building, what worked, and the results I'm not going to dress up.
What does a factory's electricity actually cost?
A data project: estimate a GB factory's day-ahead wholesale electricity exposure (~10.8 p/kWh), then prove it against what actually traded — within 0.5%. On scoping, honest uncertainty, and validation.
A leakage-free fraud feature store, built with dbt
Turning raw card-payment tables into a feature store for fraud models on dbt and BigQuery — with one rule that matters: a feature can only ever see data from before the transaction it describes.
MedGemma CKD: building a clinical RAG assistant three ways
Why I built the same chronic-kidney-disease assistant at three levels — simple, agentic, multi-agent — and the shared foundation underneath. Part 1 of the MedGemma CKD series.
From 22 clinical PDFs to a vector store
The unglamorous half of RAG: OCR with Docling, section splitting, block-aware chunking, and embedding 22 clinical guidelines into ChromaDB. Part 2 of the MedGemma CKD series.
Level 1 — Simple RAG, the honest baseline
Retrieve, answer, cite — and the retriever choices (flat, tree, RAPTOR, contextual) that decide whether any of it works. Part 3 of the MedGemma CKD series.
Let's build something
Tell me what you're working on. I'll reply within a day, and the first conversation — scoping out whether ML even helps — is free.