Capability is commoditized. Ownership is not.
Anyone can demo an agent. What you can't buy off a shelf: workers designed for your business, proof before production, decisions your reviewers can read, and code you keep.
The delivery record.
Portal and worker counts are computed from our delivery records. A digital worker is a generated, governed software worker delivered as owned source code — not a rebadged bot or script.
Workers designed to order.
No catalog to pick from. Describe your business and an AI analyst designs a digital worker around how you actually work, in up to five alternative versions. Each arrives with its steps, its approval points and an estimate of hours saved and running cost. Starter packs can pre-fill the description; the design itself is generated.
Built from your context
Your industry, team, systems, rules and goals shape every design.
Several ways to describe it
A guided ten-step questionnaire, a single paragraph, a chat assistant, uploaded documents, or an event log exported from your systems. You choose what to share, and a data-processing agreement is in place before anything moves.
A full plan, not a diagram
Every design arrives with its steps, its approval points, its connections and an estimate of hours saved and running cost.
Rules in plain English
Describe a rule in ordinary words and the Studio writes the logic, for you to accept or discard.
Rules first
A design mixes fixed rules with AI steps. A decision built as a rule is evaluated by code, not by an AI model, so it gives the same answer for the same input and adds nothing to your AI bill. Every design shows its share of fixed rules.
How the Foundation worksA worker designed for your business, with an estimated cost per run and per month that you can see before you export anything.
Proof before production.
Every worker gets its own test data: fictional records generated to match its data, or a file you choose to import. Run a scenario and watch each step, approve or reject where the worker stops for a person, and compare one version with the next, before a single real record is touched.
Tested first
Test runs are designed for made-up data. The Studio generates fictional records, so an evaluation that starts from a description of your process needs none of your data.
How the Foundation worksTest data shaped like yours
Fictional records that match the worker's own data, generated by the Studio or imported by you.
A score for every evaluation
Evaluation runs are scored for speed, consistency and coverage, and for accuracy against the checks you define.
Compare versions
Run the same scenarios on two versions and see the results side by side.
Edge-case tests
Built-in scenarios such as a missing field, an oversized input or a failure midway put the worker to the test.
You see the worker do the job, and hold a report on how it did, before you export it.
Decisions your reviewers can read.
Every worker is a readable, versioned design that your governance team can compare, test and review before it is exported. What you take away is code you own, not a dashboard you rent.
A readable design
What a worker does, who approves what and where work goes to a person, captured as a design rather than a tangle of prompts. Your engineers compare versions the way they compare code, and your governance team reads it before anything is exported.
How the Foundation worksVersioned like code
Every save is a numbered version. Compare any two to see what changed, and merge them.
Approval steps
Place an approval step wherever a person must decide. The worker stops there until someone approves or rejects.
Checks that can block an export
The Studio checks a design's structure before export. For example, a step that acts on another system must come after an approval step.
The four checks, on the Nx·Studio pageA pass mark for AI steps
Set a minimum score on steps judged by AI and choose what happens below it: stop, flag and continue, or take another path, such as review by a person.
Your reviewers read what a worker will do before it does it, and the design they read is yours.
Software you keep and can check.
The Studio delivers a complete software project, not slides: the worker's design, the code that runs it and the files to start it, with a threat model and a signed record of every file. You run it on your own infrastructure. No license check. No usage meter. No calls back to Nexgile.
See what is in every downloadConnects to your systems
Workers call any system with a web API, use tools published through the open standard for AI connections, and query the databases you allow. Credentials stay in your own vault.
A signed record of every file
Confirm, offline, that what you deployed is exactly what you were given.
People in the loop
Approval steps are part of the worker you own. It stops at every point you design for a person's decision.
Nothing stops working if you stop paying us, and you don't have to take our word for it.
Options for operations
When you export, add monitoring dashboards, alert rules, drift monitoring and a generated operating guide.
100% yours
Full source code that you host on your cloud, your private cloud or your premises, with no license keys and nothing that depends on Nexgile to run.
Ownership you can check
Every download lists what is inside it and carries a signature over every file. Your engineers run one script, offline, to confirm that nothing was altered. Because it is all source code, they can also read it to see that nothing depends on Nexgile.
How the Foundation worksThree ways to get a digital workforce.
Each lane is a real choice with real trade-offs. Here is how we'd describe them — including ours.
Build it yourself on open frameworks.
You own the code — and the years of plumbing, governance, and maintenance you'll build before the first worker ships.
Rent a hosted agent platform.
Fast start — on their runtime, behind their meter, inside their black box, for as long as you use it.
Generate it and own it.
Designed for your business, tested on made-up data, delivered as source code in your hands. This is the lane we built.
Artifacts, not adjectives.
The capabilities above produce things you can read. Two examples — both illustrative, both standing in for artifacts your team receives.
- Intake: vendor invoices from email and the ERP queue
- Check: purchase-order match, payment terms, duplicates
- − Approval: mismatches go to one approval step
- + Approval: mismatches above a limit need a second approval step
- Low scores: routed to a person
Illustrative. A policy change reviewed like code, and read before anything is exported.
How designs work, on the platform page- Run details: the worker, its version and the scenarios used
- Scores: accuracy, speed, consistency and coverage
- A pass or fail result for every scenario
- A pointer to the worker's threat model
Produced by an evaluation run on made-up data: a document you can hand to procurement.
See how evaluation works, on the Nx·Studio pageThe same capability, two very different endings.
- Their runtime.
- A meter that punishes success.
- A dashboard over someone else's black box.
- Cancel, and the workforce vanishes.
- Your infrastructure.
- Fixed rules that add nothing to your AI bill.
- A design your reviewers read before anything runs.
- Cancel, and nothing happens — it's your code.
We charge to design your workforce, not to run it. Our incentives point at your independence — ongoing help is optional, on your terms.