Machine Learning Engineer | Data Scientist | Production AI Systems Builder
π London, UK | π MSc Data Science (Distinction) | πΌ 4+ Years Production ML Experience
I'm a passionate ML Engineer who bridges the gap between cutting-edge AI research and real-world production systems. With 4+ years of hands-on experience, I specialize in building robust, scalable AI solutions that solve actual business problems.
What drives me: Creating AI systems that work reliably in production, not just in notebooks.
- Production ML Systems: End-to-end pipelines on AWS & Google Cloud
- Natural Language Processing: LLM fine-tuning, RAG systems, chatbots
- Computer Vision: Defect detection, image processing, real-time inference
- Healthcare AI: Leveraging personal patient experience for better solutions
- Multi-Agent Systems: Building sophisticated AI agent architectures
Multi-agent system built with Google ADK
- Master Agent + Policy Agent + Ticket Agent architecture
- RAG + database hybrid approach with location intelligence
- 15 comprehensive test scenarios covering real-world edge cases
- Production-ready deployment with comprehensive documentation
Tech Stack: Google ADK, ChromaDB, SQLite, Python
RAG-powered chatbot with personalized suggestions
- Combines user basket history with culinary recommendations
- Deployed on Google Vertex AI with automated CI/CD
- Significantly improved user engagement metrics
Tech Stack: LLMs, RAG, Google Cloud, GitLab CI
High-precision computer vision system for manufacturing
- 98% precision in defect detection
- 0.2s inference time per image
- Deployed on factory floor with edge computing optimization
- Reduced manual inspection time by 75%
Tech Stack: TensorFlow, OpenCV, AWS EC2, Docker
skills = {
"languages": ["Python", "R", "SQL", "C++"],
"ml_frameworks": ["TensorFlow", "PyTorch", "Scikit-learn"],
"cloud_platforms": ["AWS", "Google Cloud"],
"specializations": ["NLP", "Computer Vision", "MLOps", "Healthcare AI"],
"tools": ["Docker", "Git", "Airflow", "BigQuery"]
}π Building Production-Ready AI Systems: Creating open-source projects that demonstrate real-world AI applications
π₯ Content Creation: Developing tutorial series on building multi-agent systems with Google ADK
π₯ Healthcare AI: Leveraging my unique perspective as both an ML engineer and healthcare patient to build better medical AI solutions
π¬ Research & Innovation: Exploring the intersection of AI and healthcare, particularly in patient experience optimization
π MSc Data Science (Distinction) - Northumbria University, UK
Dissertation: Assessing Wearable/Smartphone Cues for Parkinson's Disease Treatment
π BEng Electrical & Electronic Engineering (2.1) - Ho Chi Minh City University of Technology
Thesis: GANs for Image Super-Resolution
π Key Achievements:
- 98% precision on production defect detection model
- 95% accuracy in competitor price mapping system
- 65% F1-score in personalized treatment prediction
- Successfully deployed ML models processing 1000+ sensors from 1,600+ patients
I'm always interested in discussing:
- Production ML challenges and scalable solutions
- Healthcare AI applications and ethical considerations
- Consulting opportunities for AI system development
- Open-source collaborations on production-ready projects
π§ Technical Consulting: End-to-end ML system design and implementation
π Project Collaboration: Open-source contributions and joint development
π Knowledge Sharing: Workshops on production ML and multi-agent systems
- π Website: twentytwotensors.co.uk
- π§ Email: ntg2208@gmail.com
- πΌ LinkedIn: linkedin.com/in/ntg2208
- π¦ Twitter: @ntg2208
"Building AI systems that work reliably in production and create real value for businesses and society. Every line of code should solve a real problem."
If you find my projects helpful, consider supporting my work:
π‘ "The best AI systems are the ones that work reliably when it matters most."

