Senior Software Engineer

Justin L. Sunday

Building production software and writing about AI-assisted engineering, architecture, and automation.

I work across C#/.NET, Python/Django, Angular, SQL, cloud infrastructure, and modern developer tooling, with a growing focus on what increasingly capable AI agents change about how real software gets built.

Read my writing Explore my projects About me

Writing

Browse all writing

Aug 25, 2026

Make Your Coding Agent Earn the Architecture

Evidence-Gated Capability Adoption is the workflow I use to keep coding agents from turning plausible ideas into permanent architecture before the evidence justifies it.

Aug 23, 2026

Your Parser Shouldn't Get One Shot

A practical look at resilient parser architecture, using a Reddit resume-parser discussion and lessons from a private financial application to examine redundancy, OCR, layout-aware extraction, LLM fallback, confidence, and replay.

Current areas of interest

What I write about

AI & Software Engineering

Coding agents, context engineering, automation, architecture, and what increasingly capable models mean for software development.

Software Architecture & Engineering

Practical engineering across C#/.NET, Python/Django, Angular, APIs, testing, debugging, data, cloud infrastructure, and system design.

Building Real Software

Projects, implementation decisions, failures, experiments, and lessons that come from shipping and maintaining real systems.

Building

Selected projects

View all projects

PaySpan

A self-hosted cash-flow planning system that models bills against real pay periods, reconciles statement and transaction evidence, and turns fragmented financial data into an actionable funding plan.

Python · Django · Django REST Framework · Angular · PostgreSQL · Celery · Redis

DevSculptor Jekyll Theme

A full-featured, GitHub Pages compatible, Jekyll blog theme tailored to software devs that contains project and resume pages. Contains localization support for multiple languages.

HTML · Liquid · Sass · JavaScript

Evidence-Gated Capability Adoption

Evidence-Gated Capability Adoption provides a structured way to move from promising ideas to justified engineering decisions by defining hypotheses, bounded experiments, acceptance criteria, evidence, and explicit adoption gates.

Agentic AI · Software Architecture · Experiment Design · Technical Writing

From the blog

Latest writing

Make Your Coding Agent Earn the Architecture Aug 25, 2026
Evidence-Gated Capability Adoption is the workflow I use to keep coding agents from turning plausible ideas into permanent architecture before the evidence justifies it.
Your Parser Shouldn't Get One Shot Aug 23, 2026
A practical look at resilient parser architecture, using a Reddit resume-parser discussion and lessons from a private financial application to examine redundancy, OCR, layout-aware extraction, LLM fallback, confidence, and replay.
Comparing Pixel Art Editors Jun 11, 2022
Part one of a series of posts comparing various Pixel Art Editors. Part one gives an intro to this series of posts and gives a short review of Aseprite. A comparison of various pixel art editing tools.
Building an App with Angular, Part 1: Introduction Jan 9, 2022
This post introduces the project and sets the stage for the following posts in the series. The series will cover HTTP Interceptors, logging, snackbar messages, angular material data tables, NGRx and more.
Site Updates Jan 6, 2022
No new content has been posted lately, because I've been upgrading aspects of the site.
Browse all writing