Djlan Messina

Worked on a Node.js and React app used by around 3k people daily with the backend running on AWS Lambda. Set up Python data pipelines for model training and preprocessing that fed into production ML services on AWS ECS. Automated build, test and deployment across environments using CI/CD which cut our release time by about 40%. Made some frontend tweaks around SSR and code splitting that brought page load times down by 35%. Kept production running smoothly with CloudWatch monitoring and auto-scaling hitting 99.6% uptime consistently. Worked closely with the product and research teams to turn ML experiments into actual shipped features.

Built internal analytics dashboards in Vue.js with a Python backend used daily by around 2k people. Wrote data pipelines using AWS Lambda S3 and Pandas that automated a lot of manual processing and cut batch costs by 22%. Did statistical analysis and visualisation work to help product and business teams make sense of the data. Put together a set of reusable UI and reporting components that ended up being used across 4 different product modules. Containerised the data workflows with Docker and wired up CI/CD so deployments were consistent and easy to reproduce.

Built backend services in Node.js and C# for a compliance SaaS that processed around 400k events a day on AWS. Added ML-assisted fraud detection logic into the compliance pipeline alongside existing rule-based checks. Built out React dashboards for transaction monitoring used by fintech clients across 12 markets. Set up serverless ingestion pipelines on Lambda with SQS and kept test coverage at 86% across core modules.