I got into backend work by automating the job I already had.

I started at Seneca in January 2024 in the Computer Programming & Analysis program, and the courses that stuck were the ones closest to the machine — operating systems, data structures, databases, and writing C until pointer bugs stopped being mysterious.

Partway through, I took an IT support and networking job at Jewlr, which turned out to be the most useful thing I have done for my engineering work. Supporting 120+ people across Windows, macOS and Linux means you see how systems actually fail: not in the clean way a textbook describes, but because a certificate expired, or a switch port was mislabelled, or nobody noticed a service had been down since Friday.

It also gave me a steady supply of problems worth automating. The one I keep coming back to is a Python Flask daemon I wrote to connect our Plane.so issue tracker to Slack. The interesting part was not the API calls — it was realising that a naive poller announces the entire backlog to the channel the first time it starts, so it needed to seed its cache silently and establish a baseline before saying anything. That is the kind of detail that only shows up once something is running for real.

From there the work moved further down the stack. I wrote PowerShell to automate onboarding, then started building infrastructure from code instead of clicking through consoles: Terraform for an AWS networking lab, OpenTofu and Ansible behind a Spacelift pipeline for a project where a Git push provisions a server and configures what runs on it. I passed the AWS Cloud Practitioner certification along the way, mostly to make sure the vocabulary underneath what I was building was solid.

Most recently I built IntelliApply, which is where the backend and the AI parts of my interests met. It scores how well a resume matches a job posting — and the design decision I am most attached to is that the score is computed in plain code, not by the model. The language model only explains the result it is handed. Ask an LLM to rate the same resume twice and you get two different numbers, which is fine for prose and useless for a score someone is going to act on.

In August 2026 I start a co-op term with the Ontario Ministry of Public and Business Service Delivery and Procurement, working on enterprise web development. It will be the first time I write software inside an organisation at that scale, which is the part I am looking forward to.