Corvic AICorvic AI

For developers

Build agents and apps on your company's data

Corvic gives your agents governed access to tables, documents and memory. Connect over MCP, REST API or SDK, and skip building the retrieval and permissions layer yourself.

Accounts are free. No credit card needed.

Or see the agent harness
Built forSoftware engineersAI engineersPlatform engineersSolutions engineers

The problem

Every agent needs a data layer, and you keep rebuilding it

The model call is one line. Connectors, retrieval, memory and access control are the rest of the project.

01

Connectors and retrieval by hand

Each agent needs its own loaders, chunking, vector store and sync jobs before it can answer one question.

02

Permissions are an afterthought

An agent with a service account sees everything. Scoping it to what the user is allowed to see is custom code.

03

No memory between sessions

Agents start from zero every time, or you build and host your own memory store next to the app.

What you build

Ship the agent, not the plumbing

Corvic handles data access, retrieval and memory. You build the experience on top.

Agents over MCP

Point any MCP-capable agent at a Corvic workspace and give it tools to query tables, search documents and read memory.

OutputTools your agent calls, scoped to its access

Apps on the REST API or SDK

Call Corvic from your own product to answer questions, run pipelines or fetch results.

OutputAnswers and tables in your own UI

Retrieval without the stack

Semantic search and knowledge graphs over your data, kept fresh by pipelines, without hosting a vector store.

OutputSearch and graph queries over live data

Agent memory

Give agents memory that persists across sessions and is shared across the agents you allow.

OutputContext your agents keep and reuse

Model choice per task

Pick the model per step, including open-source models or your own endpoint, and change it in config.

OutputOne agent, the right model at each step

Internal tools, fast

Generate a working web app on live tables, then extend it, instead of scaffolding an admin panel.

OutputA live web app on governed data

How it fits

Your agent, any model, governed data underneath

The Agent Harness sits between your agents and your data: it routes each request to the right model and the right sources.

Your stack

Models

  • OpenAI
  • Anthropic
  • Google
  • xAI
  • Open source

Or your own endpoint

Connectors

  • SalesforceSalesforce
  • SlackSlack
  • StripeStripe
  • GitHubGitHub
  • NotionNotion

+38 more over MCP

Your data

  • SnowflakeSnowflake
  • DatabricksDatabricks
  • AWS S3AWS S3
  • PDFs & scans
  • Audio & images

No modelling first

The agent harness

Model

Any provider

Swapped in config, not code

Tools

Over MCP

Read and write where work happens

Grounding

Your own data

Answers cite the row they came from

Access

Existing permissions

Scoped per person, per source

What you get

ChatExplore your data, with citationsKnowledge graphsHow things connect, and what they meanPlaybooksA good run, repeated on a scheduleData pipelinesThe tables underneath, kept currentAgentic data processingWhole document piles into clean rowsLive web appsPublished to a URL your team opens

No pipelines to build. Configure the harness once — swap the model, add a connector, tighten a permission, and everything above keeps working.

How it works

From your data to a working agent

Connect once. Every agent and app you build reuses the same governed layer.

  1. 1

    Connect your data

    Databases, warehouses, buckets and SaaS tools. 220+ sources, kept fresh by pipelines.

  2. 2

    Set access once

    Permissions per person and per source carry through to every agent that acts for them.

  3. 3

    Call it from your code

    Use MCP, REST API or SDK, inside or outside your organization.

Connects toPostgreSQLPostgreSQLSupabaseSupabaseStripeStripeGitHubGitHubSlackSlackLinearLinearAnthropicAnthropicOpenAIOpenAI

Built with other teams

Build on what your data team already runs

The tables, indexes and memory you call are the ones engineers and data scientists maintain.

Agent Harness

Agents on your data, with the model you choose per task.

Virtual Memory

Context your agents remember across sessions.

Data scientists

See how the indexes and graphs your agents search get built.

Security

Agents that pass a security review

Access control, encryption and model choice are handled by the platform, not left to each app.

SOC 2 Type II

Corvic holds a SOC 2 Type II report. Ask for it during your vendor review.

Private VPC deployment

Deploy Corvic in a private VPC. Tenants are isolated from each other.

Your keys, encrypted data

Customer-dedicated key management. AES-256 encryption at rest and TLS 1.2+ in transit.

Access per person and per source

Permissions are scoped per person and per source, so people and agents only see what they're cleared for.

Zero-retention model options

Choose models with zero-data-retention options where the provider supports it, and swap them in config.

SSO and removal on demand

Sign in with SSO. Corvic is GDPR and CCPA compliant, and your data can be removed on demand.

Common questions

What developers ask first

Over MCP, REST API or SDK. MCP-capable agents can use Corvic's tools directly; anything else can call the API.

Access is scoped per person and per source, so an agent works within the permissions it's given rather than a blanket service account.

No. Choose models per task, including open-source models or your own endpoint, and swap them in config without rebuilding the workflow.

Not for the data you bring into Corvic. Pipelines embed and index it, and Virtual Memory serves it to your agents.

Connect your first agent to your own data

Connect a source, point your agent at it and ask. Free to start, no credit card.

Accounts are free. No credit card needed.

Read the docs