Writing detail

I Built an AI Crew to Automate My Job Applications. Here's What I Learned.

A build log on orchestrating AI agents to automate repetitive application work while keeping human judgment in the loop.

Back to writing Read the original article

Key takeaways

Agent workflows are most useful when they remove repetition instead of pretending to replace judgment.
The system design balances automation with review points that protect quality.
The write-up shows product thinking around orchestration, not just model prompting.

This portfolio page keeps a concise internal summary while the full article remains published externally on Medium.

This build log explores an agent-based workflow for automating a frustrating but repetitive task: job applications. The article frames the system as an orchestration problem, where the real challenge is deciding what to automate fully, what to assist, and where a human should still make the final call.

It also reflects the kind of judgment the portfolio tries to foreground. The project is not presented as AI for novelty’s sake, but as a practical attempt to build leverage while keeping output quality and user control intact.

Previous What I Learned Building a Transformer Model from Scratch Next Beyond Keywords: How I Taught an AI to Recommend Books Based on Vibe