Company brains

An AI teammate in Slack, running in your own cloud.

A company brain is an AI teammate your team reaches in Slack. It runs in your cloud account with its own memory, skills and integrations, takes on bounded work, and hands anything consequential to a person for approval.

Last updated

In practice

What does working with a company brain look like?

Someone asks in a Slack thread. The teammate starts an isolated job with only the approved context, does the work, and posts the result back for review. Nothing changes in a business system until a person approves it.

ILLUSTRATIVE PRODUCT VIEW · FICTIONAL DATA · BASED ON OUR INTERNAL SKY INTERACTION PATTERN
# leadershipTEAM WORKSPACE · APPROVED MEMBERS
MO

Maya · Operations 9:14 AM

Prepare tomorrow's leadership brief. Compare the pipeline report with last week's decisions, then flag anything without an owner.

41

Foundry agent 9:14 AM

Starting an isolated job with the approved brief template, decision log, and read-only pipeline report.

LEADERSHIP BRIEF V5DECISIONS LOGPIPELINE · READ ONLY
41

Foundry agent 9:18 AM

Tomorrow's brief is ready for review.

LEADERSHIP BRIEF · 08 SEPREADY FOR REVIEW
3RISKS FLAGGED
2DECISIONS NEEDED
4OWNER GAPS
2FILES ATTACHED
REVIEW REQUIRED

Four task-owner updates are proposed. Nothing has been changed yet.

The coordination problem

Why not just give everyone an AI chatbot?

Because personal chat speeds up isolated tasks while scattering context, decisions, and outputs across private sessions. Foundry gives the company one governed way to delegate and review real work.

BEFORE · EVERYONE STARTS OVERAFTER · WORK MOVES TOGETHER
Disconnected personal AI compared with coordinated company work On the left, three people use separate chats, files, and business systems. On the right, people and an agent share one work record connected to approved systems and review. SESSION-BY-SESSION AI CUSTOMER-CONTROLLED WORK OPERATIONS FINANCE OWNER PRIVATE CHAT DOC COPY SYSTEM CONTEXT SPLINTERS · DECISIONS DISAPPEAR TEAMMATE REVIEWER SHARED WORKREVIEW READY SYSTEMS ONE OWNER · SHARED CONTEXT · VISIBLE OUTCOME

Before: everyone starts over

Operations, finance, and owners work from separate chats, document copies, and business systems. Context splinters and decisions disappear.

After: work moves together

One request, shared approved context, a named reviewer, and a visible outcome connected to the systems the company chooses.

Governed workflows

What is a governed AI workflow?

A governed AI workflow is one bounded piece of company work where an agent receives only the context and access the task needs, runs in an isolated job, and hands its output to a named person before anything consequential happens. Every run leaves a record the company keeps.

Foundation is the reusable infrastructure we are building beneath this work: security, permissions, connectors, approvals, memory, audit, release, failure handling, and evaluation. Each customer workflow still needs its own integration, acceptance evidence, monitoring, and recovery design, and the Workflow Blueprint states which parts are already verified.

01 · GOVERNED CONTEXT

Give the work the right context—not all the context.

Foundation selects skills, memory, configuration, and approved sources in a form the team can inspect. Each job receives only the relevant context and access.

  • Editable operating knowledge
  • Per-team memory and rules
  • Approved connections to business systems
CONTEXT SELECTIONCLIENT ACCOUNT
How Foundation selects context for a jobPolicies, team memory, system data, and an operating skill feed a selected context layer. The bounded context enters an isolated job, which produces a reviewable output. COMPANY SOURCESSELECTED CONTEXTBOUNDED WORK PoliciesAPPROVED FILES Team memoryCHANNEL-SCOPED System dataREAD ACCESS Workflow skillVERSIONED RULES JOB CONTEXTOnly what thistask needsACCESS CHECKED ISOLATED JOB OUTPUTReady to review NOT AN ALL-KNOWING BRAIN · AN INSPECTABLE CONTEXT LAYER

Company sources

Approved policies, team memory, system data, and a versioned workflow skill.

Selected job context

The task receives only the relevant context and access—not every available company source.

Bounded work

An isolated job completes the task and returns an output for a person to review.

02 · ISOLATED EXECUTION

Let agents work without giving them the keys to everything.

Each conversation gets a dedicated job environment. The agent receives bounded capabilities, performs the work, and reports back with an artifact or decision—not an invisible chain of autonomous actions.

  • Scoped tools and credentials
  • Separate execution environments
  • Logs and outputs the company keeps
JOB ACTIVITYJOB 041-0927

03 · HUMAN CONTROL

Put people where judgment matters.

Rules decide what the agent may prepare. Named owners handle exceptions, approve consequential actions, and keep the reasoning with the work record.

  • Explicit approval points
  • Named workflow owners
  • Reviewable actions and exceptions

Company brains

What is a company brain, and who controls it?

A company brain is a persistent context layer—decisions, conventions, documents, and operating rules—that people and agents draw on, so work does not start from zero every session. We build it in your cloud account, under your keys, and scope what each job may read.

You pay model, cloud, API, and software providers directly; Foundry does not resell usage or mark up tokens. Foundry works through delegated access you can revoke, then operates the system or hands it over. Foundation and other software rights follow the agreed license.

Talk through the architecture
  • 01
    Deployment and workflows

    Your installed environment, operating configuration, workflow definitions, and release records, subject to the agreed software license.

  • 02
    Company context

    Data, memory, configuration, and the rules that govern access.

  • 03
    Credentials and accounts

    Your provider relationships and secrets remain under your control.

  • 04
    Work records and artifacts

    Decisions, logs, files, and operational outputs stay with the company. Software rights, including code changes, follow the agreed license.

  • 05
    The option to change operators

    Revoke Foundry's role or use the runbook to operate the system yourself.

A concrete first workflow

What does a first workflow look like?

Something people already do every day. In this illustrative clinic example, a referral case moves from intake to preparation, one human decision, and a system outcome—not “chat with AI.” More healthcare workflows are on the healthcare page.

REFERRAL FOLLOW-UP · ILLUSTRATIVE

One owned case from request to outcome.

The agent prepares routine work. Staff see exceptions and approve action. The record keeps the context, draft, decision, and outcome together.

12FILES REVIEWED
8DRAFTS READY
3MISSING ITEMS
1STAFF DECISION
WORK RECORD · 041-0927OWNER · REFERRAL TEAM
01
Intake opens the case

Approved referral fields and source records enter one work record.

COMPLETE
02
Agent prepares follow-up

Packet checked, missing information flagged, drafts saved.

COMPLETE
03
Staff handles the exception

A named owner reviews the one item outside the approved rule.

REVIEW REQUIRED
04
Approved outcome writes back

Only after approval; the work record retains the decision and output.

PENDING

How a design-partner engagement works

How does an engagement work?

It starts with a short fit call and a paid, fixed-scope Workflow Blueprint. If the Blueprint confirms fit, the design-partner target is one governed workflow validated within 30 business days of Blueprint kickoff; then Foundry operates it or hands it over.

The Blueprint has standalone value even if we do not build. For an appropriately scoped engagement with access and integration readiness, each stage leaves a usable artifact in the partner’s environment. The 30-day window is a delivery target, not a general-availability promise.

Days 1–5

Workflow Blueprint

Map the cross-system workflow and exceptions, buyer-owned baseline and target, accountable owner, authoritative data, permissions, applicable requirements, autonomy and approvals, acceptance cases, failure handling, recovery, phased scope, and budget.

Days 6–20

Build

Prepare the agreed environment boundary, narrow integration scopes, governed context, and first workflow.

Days 21–30

Validate

Replay agreed cases, verify approval and failure paths, measure the target outcome, and document monitoring, release, rollback, recovery, and the operating backlog. These delivery controls are required even while the reusable product system matures.

After validation

Operate or hand over

Foundry operates the agreed scope, or transfers the runbook and operational control to the customer team.

Built from working software

Is any of this running today?

Yes, inside our own companies. Sky is Foundry 41's own company brain: it works in Slack, runs in its own AWS account, opens pull requests in our repositories, runs scheduled routines, and reads its own logs. Olivia is the instance we run for GrowLab.

Both are internal reference deployments, not customer deployments. A messaging thread starts a job in a dedicated sandbox with controlled access to code and capabilities; selected outputs—memory, code changes, files, todos, and routines—can survive a session. Broader work identity and automatic resume remain architectural direction, not a production claim.

Ask to see it work

Questions

Frequently asked questions about company brains

What is a company brain?

A company brain is a persistent context layer of decisions, conventions, documents, and operating rules that people and agents draw on, so work does not start from zero every session. We build it as an AI teammate in Slack that runs in the client's cloud account and scope what each job may read.

Where does it run?

In your cloud account, under your keys. You pay cloud, model, and API providers directly, with no usage markup from Foundry 41. Foundry 41 works through delegated access you can revoke.

What can it do?

Our internal deployment, Sky, takes requests in Slack threads, opens pull requests in our repositories, runs scheduled routines and reads its own logs. For a client, the Workflow Blueprint sets which memory, skills, integrations and approvals it gets.

What is Foundation?

Foundation is the reusable infrastructure Foundry 41 is building beneath company brains and governed workflows: security, permissions, connectors, approvals, memory, audit, release, failure handling and evaluation. Maturity varies by component, and the Workflow Blueprint states what is verified for your deployment.

Does it act on its own?

Only within limits you set. Every consequential step has a documented autonomy level, and anything that writes to a business system waits for a named person's approval unless the workflow owner has explicitly approved a narrow automatic policy.

How long does the first workflow take?

For an appropriately scoped design partner with access and integration readiness in place, the target is one governed workflow validated within 30 business days of Workflow Blueprint kickoff. It is a target, not a promise.

Is Sky a customer deployment?

No. Sky is Foundry 41's own company brain and Olivia is the instance we run for GrowLab. They are internal reference deployments, not customer deployments or evidence of customer outcomes.

Bring one operational problem

Give important work an owned place to run.

A short fit call and focused discovery confirm whether the workflow merits a paid Workflow Blueprint, which maps the context, approvals, integrations and production path.