Engineering Manager with 20+ years building AI platforms, cloud-native systems, and distributed infrastructure at scale — across fintech, retail-tech, and SaaS.

Currently at Tesco leading a cross-functional team delivering AI-powered simulation and decision support platforms that generated multi million pounds in business savings.

• LLM cost optimisation — proxy gateways, model routing, ensemble classifiers
• AI / Agentic platforms — LangGraph workflows, RAG pipelines, OpenAI APIs
• Cloud-native infrastructure — Kubernetes, Terraform, AWS / GCP / Azure
• High-throughput distributed systems — payments, retail-tech, fintech at scale

LLM Cost Guard

FastAPI proxy that classifies every LLM request by complexity and routes it to the most cost-effective model tier automatically.

Simple requests   → Haiku   ($5/M tokens)
Moderate requests → Sonnet  ($15/M tokens)
Complex requests  → Opus    ($25/M tokens)

• 90% routing accuracy on a 40-prompt diverse test set
• 54% cost reduction vs sending all requests to Opus
• ~$49K/year estimated saving at 100K requests/day
• Built with: Python · FastAPI · tiktoken · SQLite

Languages:      Python · Java · TypeScript · GoLang
AI / ML:        LLMs · Agentic AI · RAG · LangGraph · OpenAI APIs
Cloud:          AWS · GCP · Azure · Kubernetes · Terraform · Docker
Data:           Kafka · PostgreSQL · Redis · Spark · Elasticsearch
Practices:      DORA Metrics · CI/CD · DevOps · Incident Management

• Professional Certificate in Agentic AI & Applications — IITM Pravartak
• Building AI Agents with RAG and LangChain — IBM
• AWS Certified Cloud Practitioner — AWS
• PG Program in Cloud Computing — Great Lakes Executive Learning