FORTUNE-OS / ENGINEERING STORIES
Humans & machines.
The work we share, with the evidence behind it.
LAST 90 DAYS
LATITUDE / LAST 90 DAYS
Explore model telemetry Beyond the commit trail.
Captured traces7,353Engineering trace cohort
Reported tokens47BTrace usage, separate from Entire
Estimated model cost$48.92KUSD · telemetry estimate
224.6K tool calls captured
MODEL SPANSUnreported model241.9Kclaude-opus-5-5136.7Kclaude-opus-575.7K
Whole engineering project; district filters do not narrow these counts. Business agents are separate in the Observatory. Trace tokens, checkpoint reports and tool cohorts are not added together; estimated cost is not a provider invoice.
Autopilot participation
1,058Accepted PRs with known Fortune identityAI-assisted PRs
540Entire or explicit coauthor evidenceCheckpoints available
6008/6599Unique checkpoints in selected commitsReported output tokens
963.1MEntire checkpoint reports; not a billA SHARED MANIFEST
Who helped this ship?
85%
of accepted, non-dependency PRs have known agent involvement.
Autopilot1058
AI-assisted540
No known attribution290
Dependency automation43
Missing attribution is unknown. It does not mean “written entirely by a human”.
CAPTURED EVIDENCE
Inside the machine room
Input tokens593.2M
Cache read tokens566.5B
Cache write tokens4.7B
Line attribution samples4889
Reported agent lines676.1K
Estimated model costView Latitude
claude-fable-5-1gpt-6-astraclaude-opus-5-5claude-sonnet-5-5claude-opus-5claude-opus-5-5[1m]claude-sonnet-5gpt-6-lunagpt-6.1-solclaude-opus-5[1m]claude-opus-4-8claude-fable-5gpt-5.6-solgpt-5.4-codex
PR involvement, token usage and line attribution measure different things. Coverage and metric versions constrain comparisons.
The human + machine league
| # | CREW MEMBER | MERGED | REVIEWS | LINES − | AI-LINKED PRs | |
|---|---|---|---|---|---|---|
| 01 |  Anton Vinogradov@awinogradov | 389 | 1,488 | 461.7K | 853 | |
| 02 |  dsuranov@dsuranov | 110 | 255 | 35.1K | 222 | |
| 03 |  Anton Kostenko@soulko | 80 | 324 | 21.5K | 186 | |
| 04 |  Vsevolod@vsevolod-petrushchenko | 7 | 11 | 1,136 | 88 | |
| 05 |  Olga Venger@olga-venger | 120 | 324 | 34.7K | 66 | |
| 06 |  Vladimir Kuzmin@garvae | 11 | 20 | 650 | 56 | |
| 07 |  Arthur Kozulia@rovnyart | 26 | 128 | 10.4K | 31 | |
| 08 |  Claude@claude | 0 | 0 | 0 | 3 | |
| 08 |  Rahul Sethuram@rhlsthrm | 24 | 91 | 2,616 | 3 | |
| 10 |  e-krasnokutskiy@e-krasnokutskiy | 2 | 3 | 138 | 2 | |
| 10 |  KseniaPribylova@KseniaPribylova | 0 | 0 | 0 | 2 | |
| 12 |  Fortis AI@fortis-bot | 22 | 0 | 1,354 | 1 | |
| 13 |  Andrei Miroshnik@anmi | 14 | 18 | 6,335 | — | |
| 13 |  daudovik@daudovik | 0 | 0 | 0 | — | |
| 13 |  Kate @ekbainova | 4 | 0 | 190 | — | |
| 13 |  fortune-reviewer@fortune-reviewer | 0 | 6,617 | 0 | — | |
| 13 |  github-code-quality@github-code-quality | 0 | 1 | 0 | — |
FROM THE ACTUAL RUN LEDGER
Agent operations
6,240runs started
1,219runs with a PR
60human decision gates
11.1mmedian run duration
failed286
canceled1,937
completed3,987
running9
waiting_effect21
6240 metadata-only runs, joined to delivered PRs. Separate column-scoped read-only database role; prompts and run specifications excluded. A run, a PR and an accepted delivery remain separate counts.