AI/ML Engineer | Agentic AI & Generative AI | RAG | MLOps | AWS, Azure & GCP
I build Python-based machine learning and AI portfolio projects, with a focus on grounded retrieval, controlled agent workflows, model evaluation, serving, monitoring and governance.
My background combines healthcare analytics, cancer-stage NLP work, deployment rollback-risk modelling and technical training as a Microsoft Certified Trainer. I hold a master's degree in Artificial Intelligence and Data Science.
These six projects provide a focused tour of my engineering and architecture work.
| Project | What to review | Implementation status |
|---|---|---|
| Enterprise RAG & Controlled Agents | Document ingestion, TF-IDF retrieval, citations, evaluation, guardrails, FastAPI and Streamlit | Local implementation with deterministic extractive answers; GCP reference mapping |
| Predictive Maintenance MLOps | Synthetic telemetry, RandomForest training, evaluation gates, FastAPI serving and Docker | Local ML workflow; Cloud Run and Vertex AI deployment-readiness templates |
| Real-Time ML Streaming | Event validation, feature processing, inference simulation, anomaly rules, monitoring and drift checks | Local simulation with conceptual AWS service mappings |
| Healthcare Language AI | Synthetic text ingestion, NLP preprocessing, retrieval/RAG quality gates, API, dashboard and assurance | Local reference implementation; no clinical integration or hosted LLM |
| Healthcare Agentic Service Operations | Bounded agents, deterministic evaluation, CRM adapters, retry/idempotency and governance | Synthetic reference implementation; SaaS integrations are designs |
| Azure Solution Architecture | Requirements, landing zones, identity, networking, security, resilience, FinOps and transition decisions | Reference architecture and locally validated artifacts for a fictional organisation |
Portfolio projects use synthetic data and separate implemented code, simulations, reference architecture and future cloud deployment. They do not represent live NHS systems or production customer deployments. Each repository explains its own scope, setup and limitations.
- Healthcare AI Governance & Secure Research — inventory, access simulation, policy-to-evidence traceability and a read-only reviewer portal.
- Microsoft Fabric Adoption
- Azure Data & AI Modernisation
- Airline Snowflake & dbt Analytics
Development: Python, SQL, FastAPI, Streamlit, scikit-learn, automated testing and Docker.
AI/ML: machine learning, NLP, RAG, controlled agents, evaluation, monitoring and responsible AI.
Delivery: GitHub Actions, CI/CD, reproducible pipelines and cloud architecture across AWS, Azure and GCP.
Data & analytics: Microsoft Fabric, Databricks, Snowflake and Power BI; implementation depth and integration status are documented per project.
Interested in AI/ML Engineer, Machine Learning Engineer, Generative AI Engineer and cloud AI solution roles.