Technical Skills

Java, Scala, Python, HTML/CSS/Javascript, AWS, Terraform, SQL, Kubernetes, Docker, Kafka, CI/CD, Linux

Experience

Tech Stack: Java (Spring Reactive Stack), Kafka, OpenSearch, k8s, Github Actions, Terraform

Building a reactive promotions and basket evaluation engine powering Argos and Sainsbury's Groceries Online.

Rotated across four engineering teams, gaining comprehensive experience in diverse tech stacks and business domains.

Designed and built an event driven groceries delivery tracker using AWS Lambda, DynamoDB and Next.js and presented proof of concept and demo with over 200 attendees across the business.
Successfully migrated multiple high-traffic Scala microservices, handling over 10 million daily requests for Argos online and store-front, from a legacy cloud environment to Kubernetes with zero downtime.
Developed standardised Terraform modules for AWS Lambda and S3; Built fully automated CI/CD pipelines with GitHub Actions for testing, deployment and provisioning AWS resources using Terraform.
Developed new front‑end dashboard to forecast workload for a depot workforce planning tool, playing a pivotal role in the successful delivery of the MVP.
Contributed to internal engineering guide on AWS Lambda observability metrics and tracing best practices.

Worked with the ambassador team to run offer holder days, responsibilities includes sharing advice/personal experiences to prospective students/parents and setting up the event venue.
Ran multiple campus/building tours both individually and as a team.
Setup and demonstrated robotics projects for prospective students and parents.

Grade: First Class Honours

Final Year Project: Semi‑Supervised Learning with Pseudo‑labelling and Synthetic Samples. Where I investigated the effectiveness of using synthetic samples generated from a Generative Adversarial Network and pseudo‑labelling technique for the task of improving image classification accuracy on the CIFAR‑10 dataset.
Led a team project on semantic segmentation on cardiovascular MR images with a custom U‑Net architecture. Achieved Dice coefficient of 0.8.
Robot Supermarket Assistance, a group project where we designed and implemented a robot with ROS, that would build a map of the supermarket using SLAM and QR code to locate grocery aisles and traverse to the given aisle.

Final group project where we established a non fungible sharing economy token. Covered the business model, implementation and legal challenges. I was responsible for developing the Ethereum NFT (ERC-721) contract for the token.
3 week course covering the technical, economic and legal aspect of Blockchain technology. Delivered by both academics at UZH and industry members.

Placed 1st out of 7 teams.

Challenges involve utilising the Baskerville Tier 2 HPC system to speed up the data engineering pipeline with Dask dataframe, and deep learning training with multiple GPUs using Horovod.
Designed an energy/cost‑efficient HPC cluster with 15 server racks, where our design was the most efficient.
Blog: (blog.bham.ac.uk/bear/2022/07/01/bear‑challenge‑2022/)
Interview: (blog.bham.ac.uk/bear/2022/07/01/bear‑challenge‑2022‑winners‑interview/)