Summary
Lead AI Engineer in London with a track record of taking AI from
prototype to production in enterprise settings. Currently leading AI
engineering for AWTG — contributing to the KAI AI platform and leading
delivery on an education-platform project — with a career spanning
machine learning engineering, research-computing support and
production GenAI. Comfortable across the whole delivery arc: problem
definition, retrieval and RAG/GraphRAG design, agent orchestration,
response validation, guardrails, evaluation and cloud-native
operations on Azure AI Foundry and GCP.
- Based London, UK
- Current role Lead AI Engineer, AWTG Ltd
- Focus Enterprise GenAI & agents
- Status Open to conversations
Experience
Lead AI Engineer / GenAI Specialist
AWTG Ltd · London, UK
Current - Contribute to the KAI AI platform — production-grade LLM workflows, Python/FastAPI services, backend systems, data pipelines and reusable AI capabilities — supporting enterprise AI adoption for 65,000+ active users, including the British Council.
- Led AI engineering delivery for an education platform (not yet publicly launched) — translating product, pedagogy and user needs into scalable, AI-enabled platform features.
- Built RAG and document-grounded AI workflows across approved knowledge sources, retrieval logic, structured context, embeddings and vector search, prompt construction and response validation.
- Designed agent-style AI workflows and LangGraph/LangChain-style orchestration connecting LLMs with retrieval, backend APIs and tool-style service calls — multi-step logic, error handling and validated outputs.
- Improved document-intelligence workflows by processing structured and unstructured data, running data-quality checks and preparing content for retrieval.
- Strengthened safe, responsible AI delivery through prompt versioning, approved knowledge sources, guardrails, response constraints, controlled releases and mitigation of hallucination, data leakage, prompt injection and unsafe outputs.
- Reduced manual reporting effort by 40% with automated data pipelines and reporting workflows, and improved network energy efficiency by 28% across 200+ network nodes with optimisation, analytics and decision-support tooling.
- Operate cloud-native AI services on GCP and Azure AI Foundry with Docker, Kubernetes, Redis and GCP Pub/Sub — CI/CD, structured logging, debugging, root-cause analysis and cross-team knowledge sharing.
- RAG
- GraphRAG
- LangGraph
- Azure AI Foundry
- GCP
- FastAPI
AI Research and Computing Support Apprenticeship
Teesside University · United Kingdom
Sep 2022 – May 2023 - Supported AI-focused research prototypes by configuring Python workflows, reproducible environments, automation tools, testing and technical documentation.
- Improved reliability of research computing environments through Linux setup, access management, performance checks, troubleshooting and issue investigation.
- Explained software and infrastructure issues clearly to technical and non-technical users.
- Python
- Linux
- Research computing
Sep 2022 – May 2023 Machine Learning Engineer
Unifun · Kathmandu, Nepal
Oct 2017 – Oct 2020 - Delivered AI and machine-learning capabilities for product and business use cases — data preparation, model experimentation, backend integration, testing, deployment and documentation.
- Built Python-based ML and automation workflows processing structured and unstructured data, with predictive analysis and repeatable data-driven tasks.
- Built NLP workflows, recommendation-style logic, search functionality, model evaluation and integrations between AI outputs and user-facing systems.
- Improved model reliability and data quality by validating features, reviewing failure cases and documenting issues across ML-enabled features.
- Python
- NLP
- Recommenders
- MLOps
Oct 2017 – Oct 2020 Achievements
35% Automation coverage Backend logic and API integrations for orchestration and intelligent task execution.
70% Faster 5G validation Improved data-ingestion and visualisation modules for test-cycle analysis.
99.9% Service uptime CI/CD and operational practices for telecom AI and automation services.
Technical toolkit
Languages
- Python
- SQL
- JavaScript
- TypeScript
AI engineering
- LLM applications
- RAG
- Semantic retrieval
- Prompt engineering
- Prompt evaluation
- AI agents
- Response validation
- Guardrails
Models & frameworks
- OpenAI API
- Claude
- Azure AI Foundry
- LangChain
- LangGraph
- Custom GPT workflows
Product & backend
- FastAPI
- REST APIs
- Backend orchestration
- API integration
- Full-stack delivery
- React
Data & retrieval
- Embeddings
- Vector search
- Neo4j
- Data pipelines
- Reporting workflows
- Large-context systems
Cloud & delivery
- GCP
- AWS
- Azure
- Docker
- Kubernetes
- CI/CD
- Linux
- Redis
- GCP Pub/Sub
- Observability
Education
MSc Computing — Distinction
Teesside University
2021 – 2023 BSc Computing — First Class Honours
Leeds Beckett University
2018 Activities & community
- 01 Active AI, ML and data-science writer on LinkedIn — including the Reliable AI newsletter.
- 02 Member and mentor of the LangChain community in London, supporting AI agent, RAG and multi-agent architecture discussions.
- 03 Contributor to open-source AI projects and MLOps best-practice sharing at cloud and AI engineering meetups.
