❯ cat projects/pikesquares.md

PikeSquares

Employer · Eloquent Bits · 2023–25

Python backend work on an open-source developer-tooling codebase: REST APIs, PostgreSQL, background workers, and PikeSquares, a CLI for isolated development environments.

my part: Started in QA, then backend engineer on a small team. Co-built PikeSquares; profiled and rewrote slow queries; expanded test coverage.
credits: Eloquent Bits team (pikesquares is a team project)
stack: Python, Django, Flask, PostgreSQL, AWS, pytest, CI/CD, Linux, uWSGI, Caddy
links: pikesquares.com, Code

pikesquares.com home page: Building blocks for your apps, with the CLI help output and dashboard behind a knight

## context

Eloquent Bits builds open-source developer tooling. I joined in QA and moved into backend engineering, working in a shared production codebase.

## challenge

Production REST APIs supported service provisioning, background workers and runtime workflows, and some endpoints had slow PostgreSQL queries. Test coverage was thin for a codebase that provisions people's environments.

## what i built

Production REST APIs in Python (Django, Flask) on AWS, PostgreSQL data models, background processing, and service integrations. With the team, PikeSquares: a CLI that sets up isolated local development environments.

## approach & trade-offs

I profiled production traffic to find slow endpoints, then rewrote the PostgreSQL queries and relational lookups behind them. Coming from QA, I reproduced defects first and then fixed them, and expanded pytest coverage in CI so regressions got caught before release.

## outcome

API latency dropped 15% on the optimized paths, and automated test coverage went up by about 30%. I also worked on docs, deployment and code review across the codebase.

cd .. | next: Searchable Media Archive