← The Independent 100: a study

Matt Busigin

#44 in The Sensemakers, the curator’s rank and section @mbusigin Relevance to our work: 2 of 5

Bracketed tags are evidence labels: [Verified] means we read the page, file or API response ourselves, [Reported] means a source says so and we did not check the underlying thing, and [Inference] is our own reasoning.

Credit

The Independent 100 list was curated by dan (@irl_danB) and published on 2026-09-24 at independent.prose.md (announcement). The ranks and section names are his, and number 100 is his open wild-card slot. This study is ours and is not endorsed by him or by anyone on the list.

What they publicly do

Software engineer who says he wrote Cognosis, an early LLM application platform, and who now works on polycomputing.ai [Reported: X profile x.com/mbusigin]. His GitHub account is a contributor to cognosisai/platform [Verified: api.github.com/repos/cognosisai/platform/contributors]. Recent posts cover agent control loops and AI-alignment-as-institutional-alignment theory [Verified: x.com/mbusigin/status/2102076498911928671].

Public work

  • Cognosis AI Platform repo MIT last activity 2023-02-04 · 176 GitHub stars on 2026-09-24

    An application server and infrastructure for building LLM apps (workflows, agents). He is a listed contributor.

  • yaml-runner repo no license file last activity 2023-04-14 · 56 GitHub stars on 2026-09-24

    A Node CLI that executes LLM-planned actions (commands, file edits, messages) expressed as YAML.

  • Advanced-LLM-Techniques repo no license file last activity 2024-08-28 · 0 GitHub stars on 2026-09-24

    A notes repo on building LLM agents. The README is one line; contents unreviewed.

  • polycomputing.ai product no license stated

    The site titled 'PolyComputing - Enterprise AI Solutions' rendered no content when fetched.

  • X threads on nested agent loops and Glushkov other no license stated last activity 2026-09-21

    Argues that nested agent loops should run from the smartest model outward to the smallest, citing Glushkov on semantic incompleteness, and says he translated Glushkov.

Licenses of GitHub repositories are from the authenticated GitHub license API on 2026-09-25, one row per repository in evidence/licenses.jsonl (“no license file” means the API found none; “license file present, not recognised by GitHub” means GitHub could not classify it). Other licenses are as the source states. Check the license file before reusing anything.

Relevance to our work

2 of 5 lenses: agents, orchestration

He has long practice building LLM app platforms, and his current claims about outer-loop harness design touch our orchestration lane. The public code is stale (2023) and the current work is not visible.

Scored 0 to 5 against the study’s relevance rubric: 5 means public, openly licensed work squarely in our lanes that we could study or adopt now, and 1 means little connection to our work. The score is one reviewer’s judgement of how much this public work bears on our own work. It is not a ranking of people and says nothing about the person or the quality of their work.

What we would study next

  1. Read his outer-loop control thread (process/transaction/message as harness outer-loop problems) (x.com/mbusigin/status/2102076783457624556). It is a compact claim about where small or fast models belong in a harness, testable against the loops of harnesses such as NTM and OMP. Effort: small.
  2. Skim the Glushkov 'neural program calculus' document he points to (archive.org/details/neural-program-calculus-modular). He cites it as the theory behind agent-loop design. Whether he authored the translation is [Inference]. Effort: medium.

Before reusing anything: Cognosis is MIT but last pushed in 2023. yaml-runner and Advanced-LLM-Techniques have no LICENSE file (all rights reserved by default).

Sources

Overall confidence in this record: Medium. Sources were read on 2026-09-24 and every link was checked on 2026-09-25. The full record is line 44 of people.jsonl.

Limits

Not verified: What polycomputing.ai builds (the page rendered empty); the 'first LLM application platform' claim; whether the archive.org Glushkov item is his translation.

This record comes from a first-pass survey of public material by an AI agent session on 2026-09-24, edited for publication on 2026-09-25. Nothing was installed or run for it. Roles, stars and licenses change; see the study’s limits.

Request a correction

Is something about this record wrong or out of date, or would you rather not be covered? Open a correction (a GitHub issue form with this page’s address filled in). A request to remove a record is honoured without argument; accepted corrections are recorded in the changelog with the date.

Help fix it

Each opens a GitHub issue form with this page’s address filled in. You need a GitHub account.

Made by Joshua Nowak (ZestStream) with AI coding agents. Not affiliated with or funded by Jeffrey Emanuel; Joshua uses his public tools and is a paying subscriber to jeffreys-skills.md.

Source on GitHub MIT licensed