AOS8Book a demo
The Autonomous Operating System

The loop: AI agents that improve your work, and themselves.

Define, build, run and govern autonomous workflows, worked by AI agents and approved by your people. Everything on the left is an agent's proposal; nothing reaches the right until a person approves it.

Propose & improveRun & verifySScout spots ita step that keeps waiting on a personDDrafter proposespreviewed on your past workCCritic reviewsand the way back is plannedYouRejectApproveRolls outa little at a timeWatchedagainst its targetKept, or rolled backand the agents learn from itGoverned decisionyou approve or reject, or two qualified agents

An improvement moving through the loop: purple while the agents work on it, gold at your decision, green once it runs, red if you send it back.

Continuous improvement, on an infinite loop

The agents get better tooEvery turn updates each agent's report card. Agents with a good record earn more trust; their playbooks improve through the same loop.
No agent approves itselfAn agent never votes on its own proposal, its own report card or its own promotion, and its vote counts only on work it has proven it knows.
Pause it any timeOne click stops the loop. Your processes keep running, and every way back stays open.

Self-improving, under oath: every claim checked, every change on the record, every step approved. What that means

Human in the loop, where it matters. On your choice of AI provider.

A morning on AOS

Your workflows run themselves. Your people decide what matters.

AI agents do the routine work inside each workflow, from matching invoices to answering tenders. When a decision needs a person, it reaches the right one, with the evidence, the reason and the way back.

  • Worked by your own agentsSeated in the products you install, trained on how your team works, and trusted with more only as their record earns it.
  • Human in the loop, where it mattersDecisions route by role, amount and risk. Payments always reach a person.
  • Every step on the recordEverything that happens is a command, and every command leaves an event on your own record, which you can verify.
Book a demoSee AOS Platform
Good morning, MayaFreight Ops · Carrier invoices

One decision needs you. Your agents handled the rest overnight.

  • P37 carrier invoices matched to their loadsPallas, your billing agent · each with the rate confirmation and POD it checkedDone
  • P3 invoices need a second lookamounts differ from the rate confirmation · sent to your billing teamWith your team
  • ✓Friday's carrier payments verifiedrecorded and checked against the bank fileVerified
  • SLet Pallas check routine matches under $10,000?Scout's observation · 14 of 40 invoices waited over 2 h on a person · previewed on last month's workWaits on you

Approve a $12,400 payment to Red Rock Freight? It is above the billing team's $10,000 limit, so it comes to you as controller. Pallas matched it to load LD-8851.

Pallas at this level: right 47 of 50Way back: none once paidYou'll sign in again to confirm
Example screen. Names and figures are illustrative.
Why AOS

Your team is busy doing work software should do, and checking work software already did.

Copying between systems

Invoices, orders and status updates arrive by email, spreadsheet and partner feed. Someone retypes them into the system of record, every day.

Chasing approvals

Exceptions sit in inboxes. Nobody is sure who is allowed to approve, so work waits until someone notices it is late.

AI you can't hand the keys to

AI tools can do the work, but you can't see what they did or why, and you can't stop them from approving their own mistakes.

AOS Platform · published by AOS8

Six products to define, build, run and govern autonomous work.

AOS Platform is the set of products every other product is built with. Each one has its own version, its own risk class and its own dial for how far agents may act in it.

AOS Define

Say what you need

Describe the work and drop in what you have. Agents write the brief and the definition, with its checks, and you read every rule back in plain words.

AOS Build

Versions you sign

Each definition becomes a signed version, checked and published. The catalog shows what a version is built on, what it needs and what it works with.

AOS Run

One engine runs it

Installs, sign-in, screens and every command run on the same engine, with agents doing the routine work and your people taking the exceptions.

AOS Govern

People decide what matters

The approval law, every crossing where a decision is needed, and the record of each one, sealed and witnessed outside AOS.

AOS Force

Agents with a seat

Your AI workforce: each agent sits in a declared seat, under a grant with a ceiling and an end date, at a level it has earned.

AOS Improve

It gets better, with approval

The loop: observation, proposal, preview, crossing, rollout, verification. It runs at every level, and nothing lands without a decision.

How it fits together

Published once. Installed by many. Made your own.

A publisher defines a platform, its products and their apps as signed models. You install them, fill them with your own people, agents, keys and data, and may add a layer of your own. Nobody can change what a publisher signed, and no publisher can see inside your company.

Platform

A set of related products

From one publisher, with a shared core: the records, roles and agent seats its products have in common.

Product

One area of work

A signed model with its own version, risk class and dial. Freight Ops is a product. So is AOS Define.

App

Where people work

One main record, such as a load or an invoice, with its screens, its workflows and everything that acts on it.

Your layer

What you add or change

A field, a step, a screen, a whole app. Signed under your own policy, versioned and rolled back like anything else.

Publish

A publisher defines and signs. AOS8 publishes AOS Platform. Flowwolf publishes Flowwolf AOS, which is built on it. Other companies can be admitted as publishers, and a publisher sees its installers as counts, never as records.

Install

Choose from the catalog, and AOS drafts the install for you: the plan, and the roles and seats filled from your own roster. Preview how it fits your records and settings, confirm each line, decide, and go Live. Pin a version or follow the publisher's. An upgrade is the same steps, and work in flight finishes on the version it started on.

Make it yours

Extend what you installed with a layer of your own, or define your own products beside it. A layer can tighten a guard and never loosen one, and before any upgrade it is replayed against the new version so you see what no longer fits.

One path for everything.

What always holds
  • A record changes only by a command of its own app
  • Every command leaves an event on your own record
  • An agent acts only in a seat, under a grant; a person acts only in a role
  • A refusal is an answer: it names the rule, who can fix it and what to do next
What you set, level by level
  • Settings at company, platform or product level; the most specific one wins, and the page says where each value came from
  • The dial on each product, and a lower one on a workflow if you want it
  • Which optional apps are switched on
  • Who may decide what, by risk

People never have to learn the hierarchy to use it. They work in an app, and the platform and the product show as where things come from. The model, for developers →

One shape, one loop

Say a change in a sentence. AOS drafts it. You decide.

A platform, a product, an app and a workflow all read the same way: six faces, one for each product of AOS Platform. Define shows what it is, Build how it is made, Run what it is doing, Govern who decides, Improve how it gets better, and Force who works in it. Learn one page and you can read every level.

Every change takes the same path.

  • A sentence becomes a plan, which records nothing yet
  • An agent drafts the change, and it waits at a decision
  • A person your policy names decides; the agent that drafted it never votes
  • A setting lands when it is signed; a change to a product lands when it is built
  • It is written to your record, and its way back is a sentence too
  • A change to a product you installed goes into your own layer; the publisher's version stays as signed
say a change · examples
set the dial of Load lifecycle to L1
let Iris verify "certify" in Partner onboarding
switch off Tenders
# refused, with the reason, if the app is required

add a number field "dock door" to Loads
add a step "second look" after "plan" in Load lifecycle
make "tender" a crossing in Load lifecycle

# the first three are settings; the last three change the model
How it works

From a sentence to a process that runs itself, in three steps.

No developers and no months-long project. The people who know the work describe it, and the people who are accountable approve it.

1

Describe it

Say what you need and drop in what you already have: the spreadsheet, the policy, last quarter's emails.

You“Match carrier invoices to loads and pay them through our payment provider. Anything over $10,000 needs the controller.”
2

Preview it, then sign

Agents draft the definition and review each other's work. You try the version itself on a twin of your work: walk it as any role, replay it on past cases, or run it alongside today's work with no effect. What you preview is what you sign.

Agent“Replayed on last month: 37 of 40 invoices match as your team decided. Here are the three that differ.”
3

Go Live, and let it improve

The signed version runs every day. When the agents spot a better way, they propose it with the numbers behind it, you preview it, and you decide.

Agent“14 of 40 invoices waited over 2 hours on a person this month. I can handle routine matches under the limit. Want to try it for two weeks?”
AI agents

AI agents, ours and your own.

Six AOS agents define, build, review and improve your products. Your own agents do the daily work, trained on how your team works and promoted only when they've earned it. Each one has a seat, a level and a report card, and each has something it never does.

OPlans

Orchestrator

Proposes the plan for each version, reads the other agents' work and reports.

Never decides, and never approves work it planned.

DeDefines

Definer

Works out the brief with you. Asks one question at a time and turns your documents into a complete brief.

Never drafts or reviews the design it briefed.

DrDrafts

Drafter

Drafts the definition from the approved brief: the screens, the rules, the steps and the tests.

Never reviews its own draft.

CReviews

Critic

Reviews the draft, ideally on a different AI provider, and groups what it finds by cause.

Never drafts.

SObserves

Scout

Watches how your processes actually run and reports what it saw, with the numbers.

Never proposes the fix itself.

BBuilds

Builder

Writes the signed code behind what a definition can't express, such as parsing a file. Tests first, then code.

Never reviews or approves its own code.

Your own AI agents, trained on your work

You don't build them from scratch. A product declares its seats; you seat an agent in one with a grant, such as matching invoices up to a limit until the end of the quarter. It starts by suggesting, and takes on more only when its record earns it and a person agrees.

  • Every suggestion shows what the agent looked at and how sure it is
  • Agents ask you when they need an answer, with a due date
  • A grant has a ceiling and an end date, and you can stop it any time
  • Pause every agent in one click; your processes keep running
P
Pallas · invoice matchingYour agent in Freight Ops · Carrier invoices
Trust level
Preparesdrafts for a person to approve
Agreed with people
47 of 50at its current level
Changed by a person
3 of 50each one reviewed
Next level
Checksproposed; waits on a manager

Example report card. Figures are illustrative.

Continuous improvement, on an infinite loop

It gets better every week, and only with your approval.

AOS improves itself in a loop. The same loop that improves your products also improves the agents that run them, and AOS itself.

Every improvement is previewed on your past work, comes with a rehearsed way back, and waits at a crossing for a decision. Done means verified against its target after a watch window. If it misses, it is rolled back. See the loop.

1

Your products

Steps that slow your team down get proposed fixes.

“14 of 40 invoices waited over 2 hours on a person.”
2

Your agents

Agents earn a higher level and sharper playbooks as their record improves.

“Let the matching agent check routine invoices under $10,000.”
3

AOS itself

AOS is built by the same process it gives you: the same stages, roles, records and promises. If the process is good enough to build AOS, it is good enough for your product.

“Definer asks fewer, better questions after the last update.”
What changes

Less time on routine work. More confidence in every decision.

Hours

to your first process

Managers spend a few hours reviewing. The agents do the drafting.

1 list

each morning

One briefing across every product you run. Everyone sees only what waits on them, with a deadline and a reason.

0

silent waits

Every item says when it is due and who it goes to if it is missed.

100%

of actions on the record

Who did what, why, and on whose approval. Ready for audit.

For your team

Built for the people who run the business.

Operations leaders

Change how work runs without a project

  • Describe a new process or a rule change in a sentence
  • Preview any change on your own past work before you decide
  • See what is running, what is stuck and who it waits on
Finance & controllers

Approve what matters, from anywhere

  • Payments, refunds and signatures always need a person
  • Each approval shows the reason, the evidence and the way back
  • Approve on your phone, with a quick sign-in for large amounts
Front-line teams

Start the day with the work already done

  • Agents prepare each item with what they checked and how sure they are
  • Ask “why?” about anything and get a straight answer
  • Ask for a change in a sentence: “add a detention-fee step after delivery”
IT & security

Governance you can verify, not just trust

  • Your sign-in provider, roles and approval policy
  • Data rules decide what any AI provider, or any outside AI client, may ever see
  • A tamper-evident record you can verify independently
Use cases

Start with one process. Add the next in a fraction of the time.

The first product sets AOS up for your business. Every one after that reuses your people, roles, rules and connections.

Finance

Carrier & vendor invoices

Match invoices to loads, orders or receipts, flag differences, and route payments for approval.

Finance

Customer billing & payments

Build invoices from completed work, send them, and apply incoming payments to the right accounts.

Operations

Order intake from email

Read orders from emails and attachments, create them in your system, and ask when something is missing.

Partners

Partner messages & EDI

Receive tenders, status updates and invoices from trading partners, and answer them on time.

Operations

Exceptions & disputes

Open a case, gather the evidence, ask the other party, and close it when the answer arrives.

Any team

Approvals with limits

Spend, credit and pricing approvals that go to the right person by amount, with cover when someone is away.

Flowwolf AOS · built on AOS Platform

Flowwolf AOS: the autonomous back office for freight companies.

Flowwolf is a publisher. Its platform, Flowwolf AOS, is built on AOS Platform and adds what freight needs: products for brokers, carriers and shippers, and agents that speak your language of loads, lanes, rate confirmations, PODs and the partners you trade with.

A freight company installs both, and can add products of its own on top. It works alongside your TMS, accounting and carrier-pay systems, and carrier payments always need a person.

Your own layer and productsA step, a field, an app, or a product of your own that needs the loads of Freight Ops
Flowwolf AOSPublished by Flowwolf · Freight Ops, EDI Integration and more · agents trained on freight
AOS PlatformPublished by AOS8 · Define, Build, Run, Govern, Improve, Force
AOS8The engine beneath every product · one record, one approval law
See it on your loadsWhat is Flowwolf AOS?Flowwolf AOS ↗
  • Pre-built integrations
  • Nothing to replace
  • Works with the systems you already run
  • Brokers · ShippersCarrier invoicesMatched to the load, the rate confirmation and the POD; differences flagged
  • Brokers · ShippersCarrier paymentsPrepared and routed for approval by amount, then recorded
  • Brokers · CarriersOrder intake from emailOrders read from emails and attachments, with questions when something is missing
  • Brokers · Carriers · ShippersEDI tenders and status204 tenders answered, 214 status updates sent on time
  • Brokers · CarriersCustomer billingInvoices built from delivered loads and their documents
  • Brokers · CarriersCustomer paymentsPayments applied to the right invoices, with exceptions to your team
  • CarriersDriver settlementsPay worked out from completed loads, accessorials and deductions, reviewed before payroll
  • CarriersDetention and accessorialsClaimed with the timestamps and documents that back them up
  • ShippersFreight auditEvery carrier invoice checked against the contracted rate before it is paid
  • ShippersFreight accrualsOpen freight costs accrued at month end and matched as invoices arrive
Human in the loop, where it matters

Six promises AOS keeps, built into the software itself.

These are rules the engine enforces on every command. Your company can make them stricter. No setting, no publisher and no layer can make them weaker.

AI never approves its own work

Every agent's work is approved by a person or, for routine items within your limits, by two different agents that have each earned their vote on that kind of work.

Autonomy is earned

Agents start by watching and suggesting. They take on more only after they have proven themselves, and only when you agree.

Everything is on the record

Who did what, when, and on whose approval, in a record that can't be quietly changed and can be checked outside AOS.

There is always a way back

Every change comes with a rehearsed way to reverse it. If an improvement misses its target, it is rolled back.

Your data stays yours

Passwords and keys are never shown. Personal data reaches an AI provider only if your policy allows it. Export everything, any time.

The same rules for everyone

Every company runs on the same engine under the same rules, AOS8 and every publisher included.

AOS governs decisions, not just access.

Most agent platforms govern access
  • An identity for each agent
  • Which systems and data it can reach
  • Which tools it may call
  • A switch to stop it
AOS also governs decisions
  • Who may approve what, by amount and risk
  • An agent never approves its own work
  • An agent verifies only work it has proven it knows
  • A different person when your policy asks for one
  • A preview before you decide, and a rehearsed way back after
  • A record of every decision you can verify outside AOS

Access controls stop an agent from reaching the wrong system. Decision controls stop it from approving the wrong payment. How AOS governs decisions →

You set the pace

Start with AI watching. Let it do more as it proves itself.

Each agent has a report card: how often its work was right, how often a person changed it. When the numbers support it, the agent proposes taking on more, and a person decides.

How far an agent may act in a product is the lowest of four things: the level it has earned, the ceiling on its grant, the dial on the product, and the dial on the workflow if you set that lower.

Moving money, refunds and signatures stay with people at every level.

Watches
Observes the work and learns how your team does it.
Suggests
Recommends the next step. A person does it.
Prepares
Drafts the work ready for a person to approve.
Checks
Two different agents can agree on routine items, once each has earned its vote on that workflow.
Always a person
Payments, refunds, signatures and anything above your risk threshold.
Where it runs

One codebase. Two ways to run it. Your choice.

Your company lives in a cell, walled off from every other company in it. A cell runs one of two ways, and the rules, the record and every product are the same in both.

Always on · runs today
  • Processes that always run, on a database the cell runs itself
  • Steady response at a standing cost
  • Each company's rows are walled inside the database
  • A very large or regulated company can have a cell of its own
On demand · proposed, in design
  • Work runs when it arrives, on a cloud vendor's functions and tables
  • Pay for use, with a slower first response after a quiet spell
  • Each request can reach only its own company's keys
  • The vendor is in the data path, by your choice, and is named on one page: Where your data lives

A product published once installs in a cell of either kind. Both kinds are designed to produce the same record, so a company can move from one to the other and be verified. The same design has AOS propose where your work should run, from what it measured, through the same loop, with the decision yours. The design, for developers →

Fits what you have

Works alongside the systems you already use.

Bring your existing records in and keep your current systems for what you choose not to move.

Email & attachmentsSpreadsheets & CSV importsEDI partners (AS2, X12)Payment providersYour sign-in (SSO)Webhooks & open APIYour choice of AI providerYour own AI client, as tools

Single sign-on

Use your company's identity provider, with roles mapped from your groups.

Encrypted throughout

Data is encrypted at rest and in transit, with each company's data under its own key.

Your own AI client

Each product can be offered to an outside AI client as tools. The client is seated like an agent, and nothing is open until you seat one.

Leave any time

Export all your records and their history, and verify them without us.

For developers

Built so you can check the rules your AI follows.

AOS8 is a Go core on Temporal and PostgreSQL with a React app. The rules that govern agents live in a small, tested core, and the record they leave can be verified without us.

  • One engine runs every product from a signed model
  • Products talk through published contracts: built on, needs, works with
  • Flowwolf AOS, the freight platform, is built on AOS Platform
  • A public recipe lets anyone verify the record
  • Recursive self-improvement, governed: AOS is built by the same process it gives its products, and every change passes the same approval
the recipe · verify the record yourself
# every change is an event, appended in the same transaction
event  = [seq, tenant, type, entity_id, actor,
          payload, commitment, prev_hash, ts, ...]

hash   = sha256(canonical_json(event))

anchor = [seq, tenant, from_seq, to_seq, root,
          prev_hash, ts, signature, key_id, ...]
Questions

What people ask before they start

What is the difference between AOS8, AOS Platform and Flowwolf AOS?

AOS8 is the engine beneath everything, and the organisation that builds it. It publishes AOS Platform: six products (Define, Build, Run, Govern, Improve and Force) for making and running autonomous work in any industry.

Flowwolf is a logistics publisher. Its platform, Flowwolf AOS, is built on AOS Platform and adds the freight products. A freight company installs both. Other publishers can build a platform for their own industry the same way.

Do we need developers to use AOS?

No. The people who know the process describe it, and AI agents draft it. Your team reviews every rule in plain words, and previews the version on your own work, before anything runs. Developers are only needed if you want to connect a system AOS doesn't already reach.

Can we change a product we installed?

Yes, by adding a layer of your own on top of it: a field, a screen, a step after an event, a whole app, or a product of your own beside it. You never edit what the publisher signed, which is what keeps it upgradeable. Changing the publisher's own steps is possible only at points the publisher declared. A layer can tighten a guard and never loosen one.

What happens when the AI gets something wrong?

Agents propose; they don't act alone on anything that matters. Each proposal shows what the agent looked at and how confident it is, and the person or the agents reviewing it can change it. Every change has a rehearsed way back, and you can pause all agents in one click.

What stops two AI agents from approving each other's mistakes?

Two rules. An agent never approves its own work. And a verifier knows the work: every decision names a verifying seat, and an agent's vote counts only from that seat, once it has earned it on that workflow by voting in preview and matching what your people decided.

Two votes decide only when the agents differ in substance: no shared playbook and, by default, not the same AI provider. Otherwise a person decides.

How do we change how something works?

Say it in a sentence, such as “add a step for a second look after planning”. AOS turns it into a plan, an agent drafts the change, and it waits for a person your policy names. A setting lands when it is signed; a change to a product lands when it is built. Either way it is on your record, with a way back.

Which AI does it use?

The providers you choose. You can set one for the company, for a platform, for a product or for a single agent, and the most specific setting wins. A reviewing agent can use a different provider from the one that did the work. If a provider goes down, your processes keep running; only new suggestions wait.

Can our own AI assistant work with it?

Yes. Each product can be offered to an outside AI client as tools. The client is seated like any agent: every call is a command under a grant, what it may read is bounded by your data policy, and its reads are recorded. Nothing is open until you seat one.

Where does it run?

In a cell that holds many companies, each walled from the others. Today a cell is Always on: processes that always run, on a database the cell runs itself. A second kind, On demand, is proposed and in design: it runs work when it arrives on a cloud vendor's services. The rules and the record are the same in both, and a very large or regulated company can have a cell of its own.

Is our data used to train AI models or shared with other companies?

Each company's data is kept separate and encrypted under its own key. Your data policy decides which information may be sent to an AI provider at all. A publisher sees its installers as counts, never as records.

How long does it take to get started?

Your first process is usually defined in a working day of agent time, with a few hours of review from the managers who approve it. The first one takes the most effort because it sets up your people, roles and connections. Each one after that is much faster.

What does “self-improving, under oath” mean?

A witness under oath has to tell the truth, on the record, and answers for what they say. AOS8 improves itself the same way. When an agent argues for a change, every figure it uses is checked against the record before it's shown as fact. Every proposal, decision and change is written to a record that's witnessed outside AOS8, so anyone can check it later. The six promises always hold, no improvement lands until a person approves it, and if it misses its target it's rolled back.

“Under oath” is a standard we hold the software to, not a legal guarantee.

See how AOS governs decisions

How is AOS different from other AI agent platforms?

Most agent platforms help you build agents and control what they can access. AOS starts from the work: you sign a definition of how it runs, AI agents do the work inside it, and the engine decides who must approve each decision, by amount and risk. Every change can be previewed first and comes with a rehearsed way back, and the whole record can be verified outside AOS.

Can we have our own AI agents?

Yes. A product declares its seats, and you seat an agent in one with a grant. It learns from how your team does the work, starts by suggesting, earns more trust as its record improves, and is promoted only when a person agrees. You never write an agent from scratch.

Who is behind AOS8?

AOS8 is built by Flowwolf. Flowwolf AOS, Flowwolf's autonomous back office for freight companies (brokers, carriers and shippers), is built on AOS Platform, the same platform you see here. AOS Platform itself works for any operation that can be described as roles, records and approvals.

See your own process running on AOS.

Bring one process you'd like off your team's plate. We'll define it with you and show you what changes.