Custom agents for anyone.
Not another prepackaged AI role. A platform your team builds into whatever agent you actually need — no ML team, no prompt engineers, no vendor training. Anyone can contribute.
How it works
Three things you do. The agent does the rest.
You're not writing prompts, tuning models, or managing an AI ops stack. You're teaching, equipping, and scheduling — the same shape as bringing a new teammate up to speed.
Teach it what your team knows
Playbooks capture the things your team does — incident triage, weekly digests, onboarding, quarterly reviews. Anyone can write one.
Give it your tools
Every tool your team already uses — Google Workspace, GitHub, Notion, Linear, custom internal APIs. Under credentials you provide.
Let it run on its own
Routines run on the schedules you set — hourly, daily, or on natural-language triggers. Independent of any single conversation.
What Gauss actually is
Not a prepackaged AI role. A platform for your agent.
Most AI-in-Slack products come with a fixed persona — a sales agent, a support agent, a coding agent. Gauss is the opposite. You decide what your Gauss does; it learns from your team's playbooks, your tools, and how you already work.
Prepackaged products
One shape, take it or leave it.
- Fixed role, chosen by the vendor
- Same behaviour for every customer
- Your team adapts to the tool
- Priced per seat — broad rollout gets punished
Gauss
A substrate your team shapes.
- You define the role — CS, DevOps, exec support, whatever your team does
- Behaviour comes from your playbooks, instructions, and integrations
- The tool adapts to your team
- Priced per workspace — add anyone, anytime
Who this is for
AI for the team you already have.
Adopting Gauss doesn't mean hiring an ML engineer, standing up an AI ops group, or sending anyone to a prompt-engineering bootcamp. The people who know how your team works are your team — Gauss is designed around that assumption from the start.
Playbooks read like wiki pages
The people who know how your team works write them the way they'd write internal documentation — plainly, in their own voice. No prompt-engineering course, no template syntax to memorise. Gauss handles the routing.
Instructions are plain English
Set the always-on context in the dashboard — team names, jargon, style preferences, what “done” means. No special syntax. If you can write a Slack message, you can configure it.
No dedicated AI team required
Any teammate with the right role can edit. Approvals go through the roles you already use. Gauss doesn't invent a new org chart.
How we think about AI
Enhance humans. Don't replace them.
Leave the tedious work — the reads, the triage, the digests, the checks — to the agent. Keep the judgement calls for yourself, and make them faster and better because everything underneath is already sorted.
What Gauss does
Everything you'd expect. None of the vendor lock-in.
Six things the product does. The rest of the site fills in the how.
Anyone can build one.
No ML team, no prompt engineers, no vendor training. If someone on your team can write documentation, they can teach Gauss what to do.
Fully customisable.
Every agent is shaped by the knowledge, procedures, and workspace context your team teaches it. The product isn't the persona — you are.
Every integration you need.
Google Workspace, GitHub, Notion, Linear, and anything else your team already uses. Bring your own — Gauss calls them under credentials you control.
Fully audited, always.
Every action lands in an append-only log — prompts, tool calls, responses, models, cost. Export any time. No black-box AI.
Bring your own model.
Route through your own Anthropic or OpenAI account, or point Gauss at any self-hosted model that speaks the OpenAI API.
Works on its own.
Routines run on the schedules you set — daily digests, hourly checks, on-demand kickoffs. Independent of any single conversation.
How it compares
One workspace fee, not one per person.
Everyone else charges you per seat, so broad rollout is punished. Gauss bills the workspace flat; add teammates freely.
← swipe to compare →
| Product | 20 people | 50 people | 200 people |
|---|---|---|---|
Claude Team | $500/mo | $1,250/mo | $5,000/mo |
Slack AI add-on to your Slack plan | $200/mo | $500/mo | $2,000/mo |
Viktor “Hire an AI employee” | $800/mo | $2,000/mo | $8,000/mo |
Cursor Team | $800/mo | $2,000/mo | $8,000/mo |
Glean with minimums | $800/mo | $2,000/mo | $8,000/mo |
Gauss (Team) $299 flat + LLM | $299/mo | $299/mo | $299/mo |
Comparison uses competitors' publicly-listed per-seat rates as of 2026-07 and Gauss's Team tier flat fee. LLM consumption is a separate line on any AI product; Gauss shows the markup explicitly on the pricing page.
Built for the enterprise you're becoming
Compliance and control shouldn't be an upgrade.
Transparent pricing
Every markup percentage is published on the pricing page. Set spend caps, see live breakdowns, get alerts before you cross a threshold. No surprise invoices.
Enterprise-ready controls
Single sign-on, customer-managed encryption keys, and the paperwork your legal team asks for. SOC 2 Type II in progress with a Big Four auditor.
Deploy in your own cloud
Prefer to keep everything inside your own AWS or GCP account? Enterprise plans deploy a dedicated Gauss instance right there.
Ready to see Gauss in action?
Every number on the invoice is auditable against provider list rates. See the pricing page for the full breakdown.