🤖 Explain with AI

4Geeks AI Studio FAQs¶

Short answers to the questions we hear most often about AI Studio pods, plans, the pod lead and delivery. For the step-by-step, read Create a pod.

Product and model¶

What is 4Geeks AI Studio?¶

AI Studio is a software delivery service built on pods: a human 4Geeks pod lead plus a small team of AI agents working on your repository. You choose an outcome, approve a plan and review a merge request. The pod handles planning, code, tests and evidence.

Is it an AI coding tool or a service?¶

It is a service. You do not install an IDE, configure a model or write prompts. The pod runs in a managed sandbox and the pod lead owns the result.

What is a pod?¶

A pod is the unit of work, team and billing. It has a goal, a playbook, a repository, connected tools, a plan, tasks, a wallet and a human pod lead. See AI Studio pods.

What AI models does AI Studio use?¶

The platform routes work through its Private AI Gateway and labels the tiers as 4Geeks Fast and 4Geeks Strong. The inference provider is an implementation detail and is not shown in the console.

Can I choose the model?¶

No, and you do not need to. The platform selects the model for the pod and keeps it consistent. Pods do not choose a model per task.

Is AI Studio faster than a developer?¶

It removes the bottleneck of human hours for well-scoped work: the pod can plan, code and test in parallel with your team. It is not a promise of a fixed speedup, and it does not replace architecture decisions, product judgment or code review.

Pricing and plans¶

How does pricing work?¶

There are two layers: a monthly plan with a bag of tasks and pod lead hours, and a prepaid wallet for overage. See Usage, plans and wallet.

What counts as a task?¶

An atomic, reviewable unit of work with acceptance criteria: one merge request, one bugfix, one document, one migration script, one test pass. A typical project uses 3–5 tasks.

Do I pay for tokens?¶

No. Tokens are tracked for cost and support, but they are not a customer quota and never block a task.

What happens if I need more tasks than my plan includes?¶

The extra task is charged to the wallet at the plan’s overage price. If the wallet cannot cover it, the pod pauses with a budget approval and a top-up call to action.

Do failed tasks cost me?¶

No. A failed attempt and one retry are non-billable and absorbed by 4Geeks. Only delivered work is billable.

Can I change or cancel my plan?¶

The organization owner or an admin can change the plan from AI Studio. Changes are prorated and applied immediately. To cancel, contact your account team; the current period remains active.

Is there a free trial?¶

AI Studio is a paid service. Ask your account team about a pilot or a scoped first delivery; the wizard and the first estimate can be explored before you commit.

Pod lead and help¶

Who is the pod lead?¶

A 4Geeks engineer, not an AI agent. They review the scope, answer questions, unblock stalled work and decide when a delivery is ready for review.

How much pod lead time is included?¶

Starter includes 2 h/month and Pro includes 6 h/month. Lite has pod lead on demand at $75/h. Extra time is charged to the wallet at the same rate.

How do I ask for help?¶

Use Talk to my pod lead in the pod composer, or choose I am not sure, I want help in the wizard. A request needs an active plan, pod lead minutes left and a card on file.

What is the response time?¶

Each pod shows its target response time in business hours. A stalled pod is escalated to the pod lead and appears in the operator queue.

Scope and delivery¶

What can a pod deliver?¶

Features, bugfixes, documentation, audits, QA and browser E2E, dependency upgrades, database migration engineering and the first slice of a greenfield product. See the playbooks table in AI Studio pods and the detailed use cases.

What can it not deliver?¶

• Production deployment or infrastructure changes.
• Moving client data during a database migration; the pod delivers the map, scripts, shadow validation and runbook, and your DBA executes the cut-over.
• Native mobile apps, GPU workloads, hardware, a real pentest or a public preview without the preview gateway.
• Product decisions or a full architecture redesign without a pod lead and a phased scope.

Does the pod deploy to production?¶

No. It works on a branch and opens a merge request. Your team reviews and merges, and your pipeline deploys.

What happens if the brief is too big for one pod?¶

The wizard detects it. An L brief is delivered as one declared slice with later phases visible; an XL brief becomes a program with a pod lead and a phase 1 you approve first.

Can I review the code before it is merged?¶

Yes. Every delivery is a merge request with tests and evidence. The pod never pushes to your protected branch and never merges on its own unless auto-merge is explicitly allowed.

Repository and security¶

Which repositories are supported?¶

GitLab and GitHub. You need an access token with write access to the project; the token is stored encrypted in the vault and used only by the pod’s sandbox.

Is my code safe?¶

The repository stays yours. The pod clones it into an isolated sandbox, works on a branch and opens a merge request. Credentials live in the vault and are never shown to the model or committed. Ask your account team for the current data-processing and retention terms.

Will my code train a model?¶

AI Studio uses providers and endpoints configured not to retain customer code for training. The exact terms are part of the platform’s data-processing agreement; request them from your account team.

Can I audit what the pod did?¶

Yes. The pod timeline records tasks, checks, approvals and messages, and can be exported as JSON with credentials redacted.

Quality and operations¶

How do you prevent bad code from being merged?¶

With the Quality capability, a pod cannot be delivered without a green build and test run. With E2E, a browser run is also required. A waiver is possible only with a written reason and an audit entry, and the merge is still your decision. See how Review and QA work.

What happens if a pod gets stuck?¶

The pod is marked as stalled and escalated to the pod lead. The pod room explains the reason and offers the next action.

Can I pause or delete a pod?¶

Yes. You can pause a running pod and resume it later. You can delete a pod only when it is paused, done or failed. Deletion removes the pod’s plan, tasks, timeline and sandbox; the repository, tools and billing history stay.

Can I reuse a pod configuration?¶

Yes. A pod can be saved as a template with its playbook, capabilities, tools, image and repository reference. The template never stores a secret.

Next steps¶

• See the AI Studio use cases
• Create a pod step by step
• Understand usage, plans and the wallet
• Read the Private AI Gateway

Still have questions? Ask on Discord or explore the tutorials.