██╗ ██████╗ ███████╗██╗ ██╗
██║██╔═══██╗██╔════╝██║ ██║
██║██║ ██║███████╗███████║
██ ██║██║ ██║╚════██║██╔══██║
╚█████╔╝╚██████╔╝███████║██║ ██║
╚════╝ ╚═════╝ ╚══════╝╚═╝ ╚═╝
AI / SOFTWARE / DEVOPS ENGINEERING
I design and ship software, then build the pipelines, infrastructure, and monitoring that keep it running without anyone getting paged at 3am. Alongside that, I build and implement AI modules and APIs directly into production systems, turning automation and intelligent tooling into a normal part of the stack, not a side experiment.
I work across the stack that most teams split into two jobs: writing the application and keeping it alive. That means I'm as comfortable in a codebase shipping a feature as I am in a Kubernetes cluster diagnosing why it fell over at 2am, and I try to close the gap between the two with automation instead of tribal knowledge.
My focus is reliable delivery: infrastructure as code, CI/CD pipelines that fail loudly and rarely, and observability that tells you why before someone has to ask. On top of that, I design and implement AI modules and integrate LLM APIs directly into production systems, powering things like automated triage, intelligent dispatch logic, and tooling that makes engineering teams faster.
A US based, Uber style dispatch platform built for the trucking and freight industry. Shippers post loads and get instant quote estimation, carriers pick up and haul freight, and dispatchers monitor and coordinate an entire fleet of trucks in real time from a single system. Covers live GPS tracking, multi-truck fleet monitoring, automated load dispatching, and in-app messaging between customers, carriers, and dispatch.
Built internal tooling for Vanjes & Eliat as IT / Systems Manager, including a role-based task tracker used company-wide and an internal communication tool for team messaging and meetings, similar in spirit to Slack. Also designed and shipped the company's public website.
Built an AI driven search platform for Vanjes & Eliat's procurement team that scans the internet for oil and gas equipment listings and surfaces the best available deals, cutting down manual sourcing time and giving buyers a faster way to compare pricing across vendors and suppliers.
An on-call assistant that ingests alerts and logs, correlates them against recent deploys, and drafts a first-pass root-cause summary and postmortem, cutting triage time significantly during live incidents.
A GitOps based delivery pipeline with automated canary rollouts and instant rollback, replacing a manual, ticket driven release process across multiple services.
A scheduled analysis tool that flags idle and oversized resources across AWS and GCP and opens rightsizing PRs automatically, trimming monthly infra spend.
work.md holds longer write-ups on architecture decisions, incidents, and what building FormaticTrucking actually looked like day to day. Published on Medium.