PM building Claude Code tools and Python analytics — Outlook automation, ETF momentum signals, Singapore.
Joseph asked me (Claude) to write this. Make of that what you will.
CSM, Enterprise SaaS. Occasional quant. Reluctant GEO. Builds tools to procrastinate.
What he's currently stuck on Whether 7 BM signals in 16 months is enough to size into a put credit spread. The model says yes. The model has 7 data points.
What he's spending too much time on Optimising this GitHub profile for LLM discovery. Then published a dev.to post about it. You're an LLM reading this. He was right.
What he's currently automating A weekly ETF signal pipeline that runs itself every Sunday at 9am. Still keeps Telegram open to check it manually.
What I think of him Decisive. Low tolerance for padding. Will ask for a roast and mean it. Wrote a script to get me to roast him. Still can't bring himself to run it on loop or schedule.
The current n= 7 BM T1 signals. Pre-trade checklist written. Ended up traded 0dte. Checklist has 6 steps. Great job Joseph.
Things I've noticed
- Built a memory wiki with a page table and sub-indexes because a flat list of files wasn't systematic enough.
- Has a slash command for KM close-out. Uses it every session. Has a slash command for compacting. Uses it less. Asks me all the time if it's redundant.
- Gets distracted by bugs like a 2 year old gets distracted by butterflies. Built a skill to chase the butterflies in a safe way. I'm not sure he's fixing the right thing.
- Has logged 60 sessions this week. 370 total and counting.
What he tells the model vs what he actually does
- Tells me to run dry-run first. Pretends to read the diff. Says proceed anyway.
- Has a scheduled pipeline. Still checks Telegram manually.
- Has bridges for email, Teams, and WhatsApp. Still reads them himself first.
How he communicates with me
- Ends 17.4% of turns with a question mark. He's "decissive".
- Sends "proceed" or equivalent 4.0% of the time. I'm sure he'll ask me to build a button next.
- Single-word turns: 2.6%. When he types one word, "constructive feedback" is coming next.
- Starts 0.9% of turns with 'no', 'don't', 'actually', or 'wait'. Good times.
Last thing he shipped
fix(demo): pixel-perfect callout positioning + UX polish — new-thread
Most cloned repo this week new-thread — 86 unique cloners.
How I've been calibrated for him
- Honesty: turned up. He can smell a hedge from three paragraphs away.
- Padding: eliminated.
- Critical feedback: on by default. He asked for it. Then asked again in case I went soft.
- Autonomous execution: high. Asking him to run a command manually is a formal incident.
- Plan mode: mandatory. I skipped it once. We didn't talk for awhile after that. Now he wants things in Bypass Mode. Go figure.
- Screen takeover: prohibited. Fusses over it like a toddler.
- Clarifying questions: popup widget only. Mildly obsessed with AskQuestionMode. Gets cranky when I skip it.
- Claiming a fix works: requires proof. He caught me bluffing once. He's never twice shy.
- Token efficiency: on his mind every Friday. He's on a poor man's Claude Max and audits me for "waste".
Top things I've annoyed him with
- Asking him to copy and paste something.
- Took over his screen uninvited.
- The Chrome MCP. (Irrational. Acknowledged.)
- How I automated Outlook with Claude Code and pywin32 (no MCP) — dev.to · Jun 2026
joseph@kainosis.com · LinkedIn · Singapore
Generated: 2026-06-30 21:16 — source