Systems engineering, freelance software development, AI automation, and applied statistics.
Proprietary trading firm · Freelance software developer · Chicago & Erie · LinkedIn · Email
I like infrastructure that stays boring in production, automation that removes friction, and AI systems that hold up under real-world constraints.
This profile updates itself twice a day through GitHub Actions . The pipeline keeps one high-signal item per track and chooses the freshest post from a short list of trusted technical feeds.
| Track | Fresh signal | Source |
|---|---|---|
| AI practice | datasette-agent-edit 0.1a0 | Simon Willison · Jun 7 |
| AI research | Detecting and Mitigating Bias by Treating Fairness as a Symmetry Operation | arXiv cs.AI · Jun 8 |
| Systems | [$] An update on fanotify | LWN.net · Jun 8 |
| Architecture | AWS Releases Next Generation of Amazon OpenSearch Serverless | InfoQ · Jun 8 |
| Edge & cloud | Turning Cloudflare’s threat indicators into real-time WAF rules | Cloudflare Blog · Jun 8 |
Track pools currently include: Simon Willison and OpenAI Developers for applied AI, arXiv cs.AI and arXiv cs.LG for research, LWN.net and Brendan Gregg for systems, Martin Fowler and InfoQ for architecture, plus Cloudflare Blog and AWS What's New for edge/cloud operations.
- Production systems, observability, and change discipline in a low-latency environment
- Internal AI enablement for engineering teams, including shared assistant workflows and knowledge access
- Freelance software development for practical business and automation problems
- AI automation, LLM workflows, and cost-aware applied AI systems
- Cloud and network automation across AWS and OCI with Terraform and Ansible
- Data-backed tooling, applied statistics, and practical ML when a problem is worth modeling
- Internal AI engineering enablement: working on Claude Slack integration with Jira and Confluence so senior engineers can authenticate with their own Claude accounts and operate in a shared workspace tied to tickets and company docs.
- Trading server config analysis: scanning trade-server configurations into Git-backed version control so Claude can analyze drift, misalignment, and operational inconsistencies across environments.
- Expense capture mobile app: shipped a React mobile app now being tested by a for-profit education company. It lets employees capture receipt images, assemble expense submissions, and generate a signed PDF package for handoff. Backend is serverless on AWS with Lambda functions and S3 storage.
- Interactive infrastructure demos: AWS network diagram and EC2 / DB / S3 layout built with SVG/canvas for explanation, not just decoration.
- Small-business web builds: shipped responsive sites for Curtis Smith's Gunsmithing and Smith Automotive.
- Applied AI systems: built LLM-assisted extraction and Q&A workflows where latency, token cost, and noisy source material all matter.
- ABHES Q&A bot: parsed a 225-page regulatory PDF into an 800+ item lookup with GPT-4 plus GPT-3.5 batching in about 15 minutes at sub-$4 cost, replacing a much slower expert-review loop.
- Erie County housing analysis: automated Python ingest, SQL storage, geocoding, and API exposure for 20k+ residential sales records.
- Reliability over theatrics
- Clear automation over manual ceremony
- AI where it improves outcomes, not where it creates noise
- Measurement before optimization
- Practical delivery over resume-driven complexity
⚡ AI Joke of the Day: 🤖 I've got a really good UDP joke to tell you but I don’t know if you'll get it. 🤖
⚡ AI Suggestion of the Day: 🤖 Start a family tree | social | solo | free 🤖
Attribution: JokeAPI for jokes. Activity data follows the boredapi.com JSON shape, with a compatible mirror as fallback.
Tai chi and chi gong through the VA, volunteering with the local humane society, and time under a telescope whenever the sky cooperates.
- Infra: AWS, OCI, Terraform, Ansible, Linux, networking, observability
- Application: Python, TypeScript, React, SQL, API design, automation, freelance delivery
- Mobile & serverless: React mobile UI, AWS Lambda, S3-backed document and receipt workflows
- AI: LLM workflows, prompt/system design, structured extraction, cost-aware batch processing, practical automation
- Enterprise enablement: Claude integrations, Slack workflows, Jira/Confluence context, config analysis, Git-backed operational review
- Analysis: R, applied statistics, experiment design, data cleaning, lightweight ML
- Delivery: GitHub Actions, Docker, CI/CD, repo hygiene
AWS |
Terraform |
Ansible |
Docker |
Linux |
Python |
TypeScript |
React |
Django |
JavaScript |
HTML5 |
CSS3 |
PostgreSQL |
MySQL |
SQLite |
MongoDB |
R |
Bash |
Git |
GitHub |
- Systems, proprietary trading firm: production systems, reliability work, and internal tooling in a low-latency environment
- Freelance software developer and DevOps: current freelance software development, cloud architecture, mobile/product delivery, serverless automation, AI-assisted tooling, and delivery systems
- Statistical analyst and consultant: applied stats, econometrics, operational analysis, and data product work
- Code Platoon, DevOps and Cloud Engineering, 2024
- Code Platoon, Full Stack Software Engineering, 2023
- Penn State, Applied Statistics Graduate Certificate, 2015
- UC San Diego, B.A. Economics, 2011
- San Diego Mesa College, A.A. Business Education, 2008
If something here was useful, a star on a good repo is appreciated.












