Dust
Build custom AI assistants for your team
Overview
Dust lets teams build, deploy and share custom AI assistants — pull from internal docs, customer-support tools, code repos. French-headquartered with strong European enterprise adoption. Free tier; paid plans for team and enterprise.
Use cases
What you can produce with Dust
- Build a custom AI assistant grounded in your company's Notion, Google Drive, Confluence and Slack content that answers employee questions with citations to the source documents.
- Deploy an agent directly into a Slack channel so support or sales staff can query internal knowledge without leaving the conversation.
- Connect a GitHub repository and create a coding-aware agent that answers questions about your codebase and drafts technical documentation.
- Chain multiple tools and data sources into a single agent using Model Context Protocol integrations, such as pulling CRM records and drafting a follow-up email in one step.
- Set granular admin controls over which data sources each agent and team can access, with workspace-level governance for compliance review.
- Create shared, reusable agents (for example an HR policy answerer or an RFP-response drafter) that anyone in the workspace can invoke and improve.
ASEAN Perspective
Dust in Southeast Asia
ASEAN-region availability and pricing notes coming soon. Drop the editorial team a note via /contact/ if you can supply local context (Singapore/Malaysia/Indonesia/Thailand/Vietnam).
Dust is one of the stronger platforms for building custom AI agents over a company's own knowledge and tools — connect Notion, Slack, GitHub, Google Drive, Salesforce and more, then assemble model-agnostic agents (across OpenAI, Anthropic, etc.) that automate real workflows. For ops, support, engineering and internal-knowledge use cases it offers a good balance of power and approachability, and the model-agnostic stance avoids lock-in.
It is more involved than a plug-and-play chatbot: getting value means thoughtfully connecting data sources and designing agents, so there is a real (if manageable) setup curve, and it is priced for teams. Globally available and English-first, with a developer-friendly API for embedding agents. A solid choice for companies serious about operationalising AI internally rather than dabbling.
What people say
Dust has quietly become one of the more credible European entries in the crowded "AI agents for the enterprise" category. The Paris-based company, founded by ex-OpenAI engineer Stanislas Polu, raised a $40 million Series B in May 2026 led by Sequoia Capital and Abstract, with Snowflake and Datadog participating, and now positions itself as an "operating system for AI agents" rather than a chatbot builder. It adopted the Model Context Protocol early, which reviewers repeatedly cite as a sign the platform is betting on open connectivity rather than lock-in.
User feedback is strongly positive but still thin: Dust holds a 4.9/5 rating on G2, though across only around 20 reviews, so treat that number as an early signal rather than a verdict. What users consistently praise is how quickly non-technical teams can assemble useful agents grounded in company data — Slack, Notion, Google Drive and GitHub connectors get frequent mentions — and the admin-heavy governance controls that let IT decide exactly what each agent can access.
The complaints are the flip side of building fast: reviewers report interface lag and slow agent response times, occasional connector hiccups (Google Drive comes up specifically), and frustration from technical users who want a fuller API for programmatically creating and updating agents. Several reviews describe onboarding as overwhelming, with a real learning curve around which agent tools do what. Pricing is straightforward at €29 per user per month for Pro, but Enterprise requires 100+ seats, which shuts out mid-sized companies wanting SSO and regional hosting.
Dust genuinely fits organisations that live in Slack, want many teams building governed agents on shared company knowledge, and have an admin willing to own the rollout. Teams wanting a polished consumer-grade assistant out of the box, or deep API-first automation, will feel the platform's youth.
Summary of public user & expert reviews, compiled by RECATOOLS.
About this listing
This entry was compiled from publicly available data including Dust's official website, press releases, documentation, and reputable third-party publications. RECATOOLS is not affiliated with Dust unless explicitly stated.
Third-party AI tools update their pricing, features, availability, and policies frequently. Information here may be outdated by the time you read this — we make reasonable efforts to keep listings current, but cannot guarantee absolute accuracy.
For the latest details, please refer to Dust directly →
Spotted something out of date? Suggest an update →
Dust in the news
Developer Tools
OpenTelemetry Graduates CNCF: 12,000 Contributors Lock In the Industry's Unified Observabi...
AI & ML
Microsoft Credits AI With Finding Bugs in May 2026 Patch Tuesday — and the Industry Should...
ASEAN Tech
Thailand's EV Revolution: How Southeast Asia's Car Industry Is Being Reshaped
More in Productivity
