hamedmp
@hamedmpJoined Nov 2026covers Dec 6, 2025 – Nov 17, 2026
- sessions
- 546
- tokens
- 38.8B
- active days
- 169
- longest streak
- 62 days
- estimated spend
- $12,414
leaderboard ranks
earned titles
The Hands-On Pilot
1.51 mid-run corrections per session (1 or more earns it)
agents, by share of sessions
- Codex 86%
- Claude Code 7%
- Pi 4%
- OpenCode 2%
- Cursor 1%
fun factRan 13 agents at once.
how they work
ratios of counts, not volume- Mid-run corrections
- 1.51
- Sessions with a context reset
- 16%
- Tool-call error rate
- 6%
- Started with a plan
- 42%
- Sessions at once
- 1.7×
Recap cards
16 cardsPersona
The Morning Orchestrator
546 sessions in 347 days
Most agents at once
13 at once
sessions side by side, plus 629 subagents
Tokens processed
38.8B
tokens, about $12,414 at API list prices
Longest streak
62 days
in a row, Sep 17 to Nov 17
Longest session
57.1 hrs
of active work in one Codex session
When they prompt
Morning mind
busiest hour 00:00
Manners check
32 pleases
9 thank-yous and 2 sorries in 2,925 prompts
Danger zone
290 × rm -rf
11 force pushes and 2,575 naps. Nothing caught fire.
Time with agents
505 hrs
with an agent at work, 861 agent-hours in all
Favourite day
Thursday
the busiest day, 1,011 prompts
Test runs
3,342 runs
of the tests, 67% passed
Bash hall of fame
74,011
shell commands run by agents
Where things broke
10,565
failed tool calls, 6.3% of 168,179
Words typed
99.0K
words in prompts, 25% of them questions
Slash commands
69
slash commands typed
Deep thoughts
147,905
thinking blocks, the longest 11.8 min
Published by @hamedmp from their own AI Insights report. Self-reported, not verified. · see the leaderboard
Find out yours.
One command in your terminal. It reads your history on your machine and makes the recap there. Nothing is uploaded.
$ npx ai-insights@latestYour report stays on your machine. Publishing a profile like this one is optional and sends only the lines you see.