
currently @clerkchat | enabling conversational messaging through AI Substack: https://xitter.cf/t.co/Z4fmOLY7Xy
Montreal
Joined January 2024
- Tweets70
- Following157
- Followers43
- Likes156
Super cool. Worth trying whitebored.vercel.app/
same idea but uses your mac camera.
You can learn on paper, with chatgpt.
Been using my finance scribe for 6 months now.
Similar to my workout scribe, it tracks my finances over time. The main goal is consistent financial habits.
Today I gave it a journal-like UI with opus 5.5
William Bowen retweeted
Bringing back the computer we fell in love with.
See you tomorrow, link below.
@heyclicky
How I track my workouts.
1. Write workout notes
2. Ai parses notes and tracks over time.
3. Chat with my workouts using ai, has complete context on my training history and patterns.
Built to reward consistency.
Built on lovable.
Reading is important.
Articles gather dust - too long to read.
Anyone else have this problem?
⚡️ I built tryTLDR ⚡️
Highlight your article, get your TLDR in 5 secs.
Comment, I'll send you the download link
#buildinpublic
#buildinpublic log:
- Wrote drafts for internal AI info docs. Empower people to use AI.
- Call with first AI client.
- Draft AI / Trello integration + build with Dev.
- Wrote a doc on AI vs non-AI automations.
- YT video on using my prompt template.
- Started 2025 AI roadmap.
YT vid on prompt template in real-life example: youtu.be/WrNXIApR3mU
YT vid on my day: youtu.be/wHs4QkQ05q8
Build in public log:
- Tinkered with a self-improving AI agent program.
- If we build a reliable, self-improving AI, we can assert that any AI agent will meet a standard.
- Build a custom TLDR agent on ChatGPT.
- Linked below.
- AI sync with the team.
#buildinpublic
I’m building AI Agents over SMS.
⚡️Today I got my first paying customer!!⚡️
If you’re someone who:
- Is interested in AI agents
- Wants to learn how to build AI agents that deliver real value to your business.
I’m going to document this whole build, make sure to follow!
⏳ 500+ support conversations per week. 📊 3+ hours of manual analysis.
Built a simple AI program that analyses our AI support agent's performance.
Results: 12+ hours saved monthly, clear metrics on agent performance, spotted 3 key improvement areas.
🧵 What worked for me:









