🏆 WEEK 6 COMP WINNER 🏆

This week

took the whole thing!

Mayston is a research partner for UK neurophysios prepping stroke-rehab CPD. That's the kind of work where you walk in ready for the pushback or you don't walk in at all. So Mayston holds the debate for you. It won't open the discussion until it knows your current teaching position, it keeps the rival treatment schools apart so they don't blur into mush, and it hedges every claim on the GRADE evidence scale. Deepest folder in the comp.

But here's why it won. This is productionize your opinion all over again. Mayston is built on a working UK physio's decades on the frontline. His dad. The follow-up calls are logged right there in the repo. James didn't research a persona, he built his father's life work into a tool, then shipped it like a launch. Landing page, a 60 second overview, a full walkthrough, annotated diagrams.

Same lesson as last week, said a little different. YOUR expertise is the value. And when it's close to home, a son building his dad's craft into something other clinicians can actually use, you feel it. People trust that. You stay passionate because you care whether it's right. That's the whole game.

📺 60-second overview: https://youtube.com/shorts/uDZyiZ-aD8w
📺 Full walkthrough: https://youtu.be/gzHoSpZPWS0
💬 https://mayston.pages.dev
🔗 https://github.com/JamesMack05/mayston

🎯 Honorable Mentions

Six builds, no particular order, that made this so hard to call.

🇳🇱

@Arjen Stet

, Lex

Dutch worker-classification law, built for the exact enforcement wave the Belastingdienst just restarted. What got me, it refuses to cite a court ruling it hasn't verified, so no made up ECLI numbers, and it splits the sources that lead (statute, the High Council) from the ones that only signal (blogs, tax-authority posture). Every conclusion gets labeled settled, in motion, or not-yet-law. Bilingual EN/NL readme so the rest of us can follow it, plus a short intro video. This was my runner-up, and it was close.

🇦🇺

@Joshua Hubbard

, Warrant

This is productionize your opinion in its rawest form. Warrant designs AI-era assessments for Australian VET, and the sample run in the repo uses his own RTO assessment files, on his own machine. He IS the user. Sharpest scoping gate in the whole comp. His rules.md flat out says "a researcher that starts before it knows the angle is just a summarizer." Six-tier source ladder that disqualifies detector-vendor marketing by name. New name to a lot of you. Remember it. The only thing between Joshua and the top spot was distribution. No video, no landing yet. Add those and this wins outright next time.

🔗 https://github.com/hubbardjoshua9-a11y/warrant

☕

@Daniel Neuhaus

, Olivia

Yeah, last week's winner, and he went and did the exact thing that won him Week 5 all over again. Built from inside his own world. Olivia researches café location decisions, written by a regional café operator in Yass, with a regional-adjustment framework no metro model would ever spit out. Refuses to search until it knows the investigation type, and closes every job with a mandatory "what this data cannot tell you" section. Operator-grade. Landing page and live demo shipped.

🚚

@Jayden Forshee

, Catalyst (and Grain)

Two entries this week, both sharp. Catalyst researches trucking-fleet deals in a domain that's a total data swamp. Four-tier credibility ladder, a catalyst-event thesis that refuses manufactured urgency, and an output that ends with "three findings that would kill this thesis." Best part, the folder literally drives the live app, so you edit a rule and the site changes. The animated demo running side by side against a generic AI is the whole pitch in 30 seconds.

🩺

@Bas Rosario

, Connective Corpus

For people with suspected immune dysregulation who keep getting told their bloodwork is normal while they feel terrible anyway. It listens, asks the better question, tiers every claim, and never diagnoses. The reference library is 27 vetted sources, and he logged the 4 he rejected with reasons. And the dedication, to his wife and daughter living with Ehlers-Danlos, tells you exactly why this one is built with that much care. Close to home.

📦

@Virgilio Robinson

, AduanIA

You know I can't run a winner post without Virgilio on it. Consistent quality, every single week. AduanIA is an eight-agent HTS customs-classification pipeline that actually reasons through the law. It sequences the GRI rules in order, ranks Section Notes over Chapter Notes over Explanatory Notes, refuses to invent ruling numbers, and ships with 99 verbatim chapter-note files. Roughly $3 a classification against a $1k to $3k attorney memo.

🎖 Worth Your Click

Six more I want to name. Different angles, all worth a click.

🧭

@Crae Säkkinen

n, AEO/GEO Researcher

Deepest architecture in the field. A router plus five specialists plus an adversarial audit stage where "a clean audit with zero findings is a failed audit."

🧾

@Greg Faysash

, VERA

Evidence tiers and funding-conflict flags for cancer patients, packed into a six-page PDF built to fit a 12-minute oncology visit. Logs its own searches so it can't fake a citation.

🏅

@Xavier Vincent

, Athena

Sports-science claims for masters athletes. Best move, it treats the absence of data for 40 to 60 year olds as the finding itself, and it never hands you a verdict, only the evidence map.

🔗 https://github.com/Xavier-XVI/ATHENA

🤝

@Nicolas Patron Uriburu

u, second-witness

Due diligence on AI-vendor claims. Caps confidence on anything a vendor is the only witness for, and goes hunting for the independent source before it trusts a number.

🔐

@Anthony Conder

, Quaesitor

CMMC readiness for small defense contractors. Forks FCI vs CUI before anything else, and the live demo refuses a controlled-data paste cold ("a general-purpose AI endpoint is not a FedRAMP-authorized environment").

📍

@Ryan Barry

, Local Pack Researcher

Competitive intel for the Google local pack. It's honest that it can't see your live pack, so it tells you exactly what to screenshot. No hallucinated rankings. And it names proximity as a confounder before it credits reviews.

A few broader notes

The names keep repeating. Half this leaderboard is people who showed up last week and the week before. Multi-week portfolios of shipped builds with video walkthroughs and brand pages are going to close deals and land jobs on their own in six months. Keep stacking the evidence!

22 comments