Ex-dev → lawyer → back to dev. Founder of Million Opportunities. Helping founders find business ideas from real market signals. Work as sadhana, code as craft.
Just entered Meta's AI hackathon, which also gives $20k for being most viral on X. I've got no following on X, let's say the $20k goodbye. But if you're a tax preparer, please beta test the tool. It reads foreign bank statements and finds the FBAR number. 3-minute demo:
MikeOSS 2.0 is here: a free open-source AI harness for legal work which you can self-host, connect to any model, or run fully local.
What’s new in 2.0:
- Integrated Drafting Interface (IDE): a Cursor for legal UI with a file explorer, document viewer/editor & chat side by side
- Word add-in + workflows in Word
- Subagents for big reviews
- Web search
- Memory (account and project scoped)
- Side chat
- Slash commands
- Orgs & audit logs
- Wider model provider support: Azure, Bedrock, @grok, @vercel AI Gateway, @MistralAI, @OpenRouter, @opencode Go and more
- Local AI with Ollama integration
Website: mikeoss.com (link to github on main page)
Thank you to all contributors, including @solirvine@Pabz@b1rdmania@Diya_555696@BitGrateful@SadhakaDev@b1rdmania@antoinelouis_@snowyukitty
New in v1.2 of my open-source growth playbook (an AI agent runs it on your codebase):
• a stage check, so no scale tactics before anyone sticks around
• a first-users strategy
• @gregisenberg's AI-directory play from this thread
github.com/fayerman-source/s…
Greg, your plugin play from this thread is now Strategy 10 in an open-source growth playbook that an AI agent runs on your codebase.
v1.2 also adds a stage check, so it won't push scale tactics before anyone sticks around, and a first-users strategy.
github.com/fayerman-source/s…
I literally can't think of a better risk/reward bet right now than building a ChatGPT plugin and getting in front of their 1.2 billion weekly users.
As of yesterday, ChatGPT recommends plugins in the middle of conversations. Millions of questions get asked every day with no plugin to recommend. Each one is an empty slot.
Facebook apps in 2007. iPhone apps in 2008. TikTok in 2019. Every big platform has a few months where distribution is free, and most people miss it.
I saw this firsthand when I lived in SF building venture backed companies. They are sacred times!
If I were starting this week:
1. Find a tool or workflow people already Google for or do by hand every week (PDF to Excel, invoice generator, weekly client reports)
2. Build it as a ChatGPT plugin
3. Write the description using the exact phrases people type into ChatGPT
4. Ship 5 of them and see which ones get picked up
5. Double down on the one with the most traction
In 12 months these slots will be crowded but right now they're pretty empty.
Most people will realize ChatGPT plugins were the biggest free distribution channel of 2026 sometime in 2027.
I scanned "Small CPA firm doing monthly bookkeeping and reconciliation for 20-30 small business clients..." on 1mil.app and this came up:
A monthly subscription service.
~15h to build. Score: 6.8/10
1mil.app/share/pick/72a8c985…
Yoga teaches how to shed personalities, attachments, likes and dislikes. How to practice yoga online? What's the best way to shed all that's been accumulated by algorithms, or is unplugging the only way?
Someone who doesn't run at all and covers zero miles a month asked me if she's got enough time to train for a marathon scheduled for September, which is next month. 😅
A ranked list can't give you conviction. Five buyer conversations can.
The tool's job is telling you which idea deserves those five conversations, and which nine don't.
The part that surprised me: for months I felt guilty about collecting ideas and pursuing none.
The data says my gut was calibrated. There was no obvious winner in the pile. "Can't choose" was the correct output.
So I rebuilt the scoring to price in competition, distribution, moat, and build cost, relative to the specific founder asking.
The new model is stingy because reality is stingy.
The scanner didn't start honest. In June I took my 6 top-scored ideas and ran willingness-to-pay checks.
5 of 6 failed.
The old score measured demand and nothing else. Demand without a path past competitors is a trap.