I manage judicial operations at Santa Barbara Superior Court. Ten years of studying how systems serve people, spotting where they fall short of their potential, and improving them from the inside.
I started building AI tools in July 2025 and I've shipped in public ever since: seventeen hackathons, five wins, three products live on their own domains. I use AI as a thinking partner, direct and validate what it produces, and ship things that hold up. Growth is measurable. Context shouldn't evaporate. Invisible problems deserve visible solutions.
Off the clock I speak to court leaders about AI adoption (NACM 2026, Los Angeles Superior Court). I'm also a Supernote devotee, a Lakers lifer, and firmly convinced that earl grey hot is the correct choice.
📂 Portfolio → Clew Labs · ✍️ Writing → dev.to
Your website has a second audience now: AI agents. Agentis Lux shows what retrieval agents experience when they read your site. Paste a URL or an OpenAPI spec, get a unified report across six categories for frontend and API. Findings, not fixes. Methodology published in full.
🌐 Live: agentislux.io · 🎥 Demo video · 📦 Repo: perseus-clew
Full CI (lint, type-check, tests against DynamoDB Local, build, audit), a deploy pipeline that smoke-tests the live health endpoint, and SSRF hardening that re-validates on every redirect hop. Built solo for H0: Hack the Zero Stack, June 2026.
Named for the thread Ariadne gave Theseus to navigate the labyrinth and find his way back.
Tools for AI-native builders who move fast and need infrastructure that keeps up. Each tool solves one problem completely. They're designed to work together, but they don't have to.
🪬 Janus Clew · Developer Growth Tracking
You've shipped three projects in three months. You're definitely better, but can you prove it? Janus reads your code, measures complexity objectively, and exports proof you can share. Not vibes. Not self-reporting. Evidence.
🏆 DoraHacks AWS Global Vibe Hackathon 2025 · Winner
🧵 Ariadne Clew · Reasoning Preservation
You build fast. You iterate messy. You lose track of why you made choices. Ariadne turns chaotic LLM chat sessions into structured clarity, capturing the aha moments, design tradeoffs, and code decisions that disappear when you move too fast for version control. Works with Claude, ChatGPT, DeepSeek, whatever you use.
139 commits, 17 test files, CI, mypy, pre-commit. AgentCore integration with no mocks.
🧠 Memoria Clew · Research Memory
You save a link while debugging. Later the link is still there, but the reason you saved it is gone. Bookmarks don't remember why. Memoria does: local-first research memory that resurfaces what you saved when it matters again, with transparent reasoning. No black-box magic.
🏆 AI Vibe Coding Hackathon 2026 · Featherless.ai 1st Place + ElevenLabs Best Project · 4th of 250
🧭 Clew Directive · AI Learning Paths
AI education is overwhelming: thousands of courses, most behind paywalls, few personalized. Four questions. A personalized path from verified free resources. A PDF you keep forever. No signup. No data stored. Multi-agent system (Scout + Navigator + Curator) on AWS serverless.
🌐 Live: clewdirective.com · 🏅 AWS 10,000 AIdeas Competition · Semi-Finalist
🔮 Metis Clew · AI Code Explainer
When you're learning to code, you hit a wall: the code doesn't feel clear, and pasting snippets into AI tools gets you paragraphs that don't match your question. Metis shows you WHAT it does, HOW it works, and WHY it was built that way. Validated through user research with career switchers. MVP shipped in 5 weeks.
🌐 Live app → metis-clew.vercel.app
🕯️ Lumen Clew · Code Health Scanner
Most code feedback comes with judgment baked in. Lumen doesn't. Paste any public GitHub repo URL and get insights across code quality, dependency health, secret detection, and accessibility, with Claude translating findings into plain English. Awareness, not shame. Reflection, not prescription.
🌐 Live app → lumenclew.lovable.app
🪶 Hermes Clew · Agent Readiness Scanner
The proof of concept that became Agentis Lux. Hermes scans your HTML/JSX/TSX files and tells you exactly what AI agents can't parse, scored 0-100. Lives in GitLab Duo Chat.
GitLab Duo Agent Platform Challenge 2026
🛡️ Anton Clew · AI Workflow Guardrails
AI coding tools are powerful, but they aren't inherently safe. Anton is a policy-driven CLI that classifies file changes as DENY, NEEDS_APPROVAL, or ALLOW before they land. Deterministic rules. No probabilistic guesses. No hallucinations.
GitHub Copilot CLI Challenge 2026
🧾 README Clew · Documentation Receipts
Audits your README against your code. Documentation is a contract, code is evidence, truth in sync. Yes, I run it on my own repos. It finds things.
Replit Buildathon 2026
🎫 TicketGlass · IT Support Transparency (Iris Clew)
IT support tickets are a black box. TicketGlass gives customers visibility into what's happening, why, and whether the same issue will return. AI-powered explanations, smart solution tracking.
SuperHack 2025
| Project | One-liner |
|---|---|
| Athena Clew | Autonomous debugging agent on Gemini 3 that resolves issues without babysitting |
| Argus Clew | Agent-readiness review on every AWS commit via CodePipeline |
| Clew Quest | Open source project discovery matched to your learning goals (Algolia Agent Studio) |
| Vigil Crest | Telegram correspondent that watches the DEV challenge feed and hedges honestly |
| Clew Forge | Lightweight framework for shipping fast without losing clarity. Start with 3 files |
Work at the intersection of courts, data, and people who deserve better tools.
| Project | What it does |
|---|---|
| Themis Lex 🌐 live: themislex.org | Plain-English AI readiness check for California court staff: what AI can help with, what it must never touch, as a PDF you can hand your supervisor |
| Permission to Approach the Firewall | AI-powered cybersecurity awareness program for court staff: phishing scenarios, training materials, and assessment tools |
| 12 Angry Rows | 11 years of jury utilization data, benchmarked against NCSC standards |
| Write of Access | AWS Textract on public court forms, with honest documentation of what it can't do |
| Hung Jury | Explored automating Crystal Reports PDFs. The blocker was data quality, not Python. Sometimes the finding is the deliverable |
| Panoptes Scout | First-pass data quality briefing agent for tabular datasets. Google AI Agents capstone |
Steep 🫖 · Your repo's fortune, steeped in truth. Tea leaf reading for code: deterministic symbols, AI narration. DEV April Fools 2026, Best Use of Google AI.
Sprint-Kit · PM skills for middle schoolers. AI scaffolds task breakdown. Zero data collection.
Proof of Scoville 🌶️ · On-chain salsa tracker built for Proof of Concept 2025. Yes, really. No regrets.
Office Keys · Gamified typing practice for workforce digital literacy, because some people are still getting left behind by something as basic as a keyboard.
Bench Power: Lakers 🏀 · Data storytelling on bench impact. Lakers forever, that's non-negotiable.
| What | Details | |
|---|---|---|
| 🥇 | DoraHacks AWS Global Vibe Hackathon (2025) | Winner · Draper University scholarship · AWS credits (Janus Clew) |
| 🥇 | AI Vibe Coding Hackathon (2026) | Featherless.ai 1st Place + ElevenLabs Best Project · 4th of 250 (Memoria Clew) |
| 🥇 | DEV WeCoded Challenge (2026) | Winner |
| 🥇 | DEV April Fools Challenge (2026) | Best Use of Google AI (Steep) |
| 🏅 | AWS 10,000 AIdeas Competition (2026) | Semi-Finalist (Clew Directive) |
| 🥈 | CCC AI Summer Camp, Cal Poly DxHub/AWS (2025) | 2nd of 14 teams · selected 1 of 100 from 1,300 applicants |
| 🎤 | NACM 2026 Midyear Conference | Presenter: "Building Bridges: How Leaders Can Enable Staff Innovation" + AI learning paths session |
| 🎤 | Los Angeles Superior Court (2026) | Invited speaker, AI adoption for court professionals |
I don't write code from scratch. I provide creative direction and validation: AI tools generate, I compare outputs, pick what works, and stitch them into something that holds up. That's a deliberate workflow, not a limitation.
Occam's Razor · Solve one problem completely before adding the next. The simplest solution that ships beats the ambitious one that crashes.
Single Responsibility · Every component does one thing well. No Swiss Army knives.
Foundation First · Live integrations over mocked features. Production architecture over quick hacks. This is always a v1, not a v-final.
Operational Thinking · Ten years in judicial systems means I know implementations have consequences. That instinct doesn't turn off.
Built with ☕, stubbornness, and a rubber duck from AI camp.
📍 Santa Barbara, CA · 🖋️ Supernote devotee · 🏀 Lakers forever
AI assisted. Human approved. Powered by NLP. All mistakes are mine.