Amin
@aminJoined Nov 2026covers Oct 1, 2026 – Nov 17, 2026
- sessions
- 80
- tokens
- 6.6B
- active days
- 26
- longest streak
- 9 days
- estimated spend
- $2,194
leaderboard ranks
earned titles
The Smooth Operator
2% of tool calls failed (3% or fewer earns it)
agents, by share of sessions
- Claude Code 98%
- Codex 3%
fun factRan 5 agents at once.
how they work
ratios of counts, not volume- Mid-run corrections
- 0.81
- Sessions with a context reset
- 25%
- Tool-call error rate
- 2%
- Started with a plan
- 27%
- Sessions at once
- 1.4×
Recap cards
16 cardsPersona
The Night Orchestrator
80 sessions in 48 days
Most agents at once
5 at once
sessions side by side, plus 47 subagents
Tokens processed
6.6B
tokens, about $2,194 at API list prices
Longest streak
9 days
in a row, Nov 9 to Nov 17
Longest session
7.5 hrs
of active work in one Claude Code session
When they prompt
Night owl
busiest hour 16:00
Manners check
5 pleases
1 thank-you and 0 sorries in 307 prompts
Danger zone
54 × rm -rf
1 force push and 798 naps. Nothing caught fire.
Time with agents
69 hrs
with an agent at work, 97 agent-hours in all
Top model
Claude Opus 5.5
answered 100% of 17,126 model requests
Favourite day
Monday
the busiest day, 75 prompts
Test runs
981 runs
of the tests, 99% passed
Bash hall of fame
15,091
shell commands run by agents
Where things broke
421
failed tool calls, 1.8% of 23,753
Words typed
9.5K
words in prompts, 13% of them questions
Deep thoughts
13,403
thinking blocks, the longest 9.4 min
Published by @amin 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.