Agents that work from what your company already knows, in its documents, its numbers and its tools.
Documents, customers and projects are indexed by meaning, and every answer links back to the documents it came from.
Sales, inventory or support, the agent can pull the numbers you ask for and put them in context.
Emails, tickets or updates are prepared by the agent and reviewed by a person before anything goes out.
The agent inherits the permissions of whoever asks, with a verified identity and role-based access.
An open model on your server or a provider’s API, chosen by how sensitive the data is, the quality you need and the cost.
Email, drive, CRM or ERP, the agent works inside the tools your company already uses.
The building blocks the major AI players run in production, chosen one by one and sized for your business.
Identify
Accounts and single sign-on, with the directory you already have.
Build
An interface to compose your agents and workflows, without going through us.
Run
Agents in code, durable execution, channels and human approval.
Orchestrate
An agent’s steps, its retries and its calls to other agents.
Authorize
Who can access what, document by document.
Enforce rules
Company rules, applied to every action.
Route
A single gateway to the model, whether it runs on your servers or behind a provider’s API.
Remember
Company memory, with documents linked by meaning.
Act
The agent uses your teams’ email, drive and CRM.
Reason
Open on your server or behind an API, depending on how sensitive your data is.
A project that leaves you freer than when you started. The hosting, the code and the know-how to evolve them stay in your hands.
Scope, priorities and deliverables agreed with you before the first line of code.
Custom-built on proven open source technology, with regular progress reviews.
You receive the code, documentation and access, with no dependency on our tools.
You learn to evolve the product, with or without us.
Recommending a technology means helping it move forward. The studio publishes its own tools and contributes to the open source projects its clients’ products rely on.
agent-memory · ai-agents · cognee
An internal knowledge base that AI agents can read. Works with Cognee or any memory engine.
git clone https://github.com/nualt/agent-memory-starterAll our public repositories, MIT-licensed and free for anyone to use.
The studio writes up its projects, from technical decisions to the limits it hit in production, so you can start yours a step ahead.
ChatGPT remembers you, Claude has its projects, Copilot reads your Microsoft files. Every vendor now sells its own memory, and each of these memories stays with its vendor. At Nualt, we set up a single memory outside the tools, which all our agents read and feed, on a server of our own.
AI & agentic workflows
Eve, Vercel's new open-source AI agent framework, promises direct integration into existing Next.js projects. Between the open-source argument and the reality of its default infrastructure, here is what to understand before adding it to your website, with an example from e-commerce.