Agentic AI platform for enterprises
Your ERP, core platforms and in-house applications stay where they are. Terrapin's agents work on top of them, with each user's own permissions, a person's approval before risky actions and a full record of every step.
“Send reminders to every customer more than 60 days overdue.”
The problem
The demo answers questions from documents. The value is in reading orders, updating records and starting processes in the systems that run the business, and that is where most programmes stop.
Reach
To act in the ERP or a custom app, a bot needs credentials and broad access. Security says no, and the pilot never leaves the sandbox.
With TerrapinSystems become governed business capabilities. Agents use them without ever holding a key or seeing an endpoint.
Sprawl
Sales, service and finance each buy or build an assistant with its own prompts, keys and vendor. Nobody can see what runs where.
With TerrapinOne platform and many Workspaces: shared tools and knowledge, with each team's context and policy kept apart.
Control
An agent sends a message or changes a record. Nobody approved it, and nobody can show afterwards how it decided.
With TerrapinApprovals bound to the exact action, and a trace of every step, policy check, tool call and cost.
Security
“Never show another customer's data” in a prompt can be talked around. Limits that matter have to sit outside the model.
With TerrapinIdentity, data scope and hard limits are enforced by the platform on every step, whatever the model says.
The solution · how it works
Each part of the platform answers one question, so responsibilities never blur. Select a layer to see what it does and where it sits in the collections request from the top of the page.
Connect anything you already run
Tool Studio wraps an API, a workflow or a database procedure as a named business capability with a schema, a risk level and a policy. The agent sees the capability. The platform keeps the endpoint and the credentials.
Governed by design
Who the agent acts as
Try it: approve, reject, or change the action after it was requested.
Channels
Publish an agent to any channel from one builder. Identity checks, rate limits and policies follow it, and rich replies fall back to plain text so nothing is lost. Delivery is phased; labels show when each channel arrives.
Use cases by team
Solution Packs are ready-made agents, tools and channels you install and adapt. The enterprise examples come from the platform specification's worked scenarios.
Is Terrapin Agentic AI right for you?
Twelve situations we hear from IT, operations and business heads. Tap the ones that sound familiar. If three or more do, talk to us about a pilot.
Tool Studio turns your systems into governed capabilities that agents can use without holding the keys.
One platform with a Workspace per team: shared tools and knowledge, separate context and policy.
Agents act as each user, inside the access that user already has. Secrets stay in a vault.
Risky actions wait for a named person, with the approval bound to the exact action.
Every step, policy check, tool call, approval and cost goes into a tamper-evident trace.
Identity, data scope and hard limits are enforced by the platform on every step, whatever the model says.
Agents read records, update them and start processes, such as overdue reminders in the ERP.
Publish one agent to many channels, with the same identity checks and rules everywhere.
Region-pinned storage and inference, per-tenant keys, and providers barred from training on your data.
Model providers are allow-listed per tenant, with daily cost caps for each Workspace.
Live takeover with an AI summary, a ticket or a call back, whichever fits the moment.
Multilingual agents with cross-language search, starting with these three.
0 of 12 sound familiar so far.
Want ready-made agents for marketing or sales? They come with Terrapin Digital Platform and Terrapin CRM. Terrapin Agentic AI is for putting agents on your own systems and processes.
Security, compliance and operations
Security is part of the design, not a setting. These controls are in the platform specification, and the figures below are design targets for the first release.
Delivery
The platform is specified and being delivered in phases. Durations are indicative and count from programme kick-off. Phase 1 is complete when two pilot customers are live.
Pilot and design partners
Pilot customers work with the product team on a real use case through Phase 1. Already in use today: Terrapin's Digital Marketeer, Digital Receptionist and Customer Service digital workers, which move onto the platform as Solution Packs.
How to buy
You always know what happens next, who owns it and what you get at each step. Start with step one or talk to us today.
FAQ
Tap a question to see the answer. Can't find yours?
A platform for putting governed AI agents to work on the ERP, core platforms and in-house applications you already run. Agents act with each user's permissions, wait for a person's approval before risky actions, and leave a full record of every step.
The platform is specified and being delivered in phases, with pilot places open now. Phase 1 brings the first agents into production and completes when two pilot customers are live. Terrapin's Digital Marketeer, Digital Receptionist and Customer Service digital workers are in use today and move onto the platform as Solution Packs.
No. Your systems stay where they are. Agents work on top of them through governed capabilities, so nothing is ripped out.
REST and OpenAPI services and your existing workflow engine come first. GraphQL, MCP servers and reviewed functions follow, then allow-listed database procedures. There are presets for OAuth2, API keys, mTLS, SAP and Oracle.
No. The agent sees a named business capability. The platform keeps the endpoint and the credentials, with secrets in a KMS-backed vault that are rotated and never placed in a prompt.
Each agent acts as a defined identity, approvals are bound to the exact action with four-eyes and MFA for critical steps, hard limits sit outside the model, and agents must pass required tests before they are published.
Yes. Every step, policy decision, tool call, approval, token and cost is written to an append-only, hash-chained log. Audit is kept from 13 months up to seven years.
Chat, forms, API, webhooks, schedules and events come first. WhatsApp and email follow, then voice and a mobile SDK. Labels on the page show when each channel arrives.
Model providers are allow-listed per tenant and barred from training on your data. You can redact personal data before model calls and pin storage and inference to a region.
Row-level security in the database, per-tenant encryption keys, and a separate vault for secrets. Separate connections for development, test and production keep tests away from live data.
English, Hindi and Gujarati first, with search that works across languages.
Bring one process and one system. In a working session we map it, the controls it needs and how an agent would run it. Pilot customers then work with our product team on that use case through Phase 1.
In a working session we map one of your processes, the systems it touches and the controls it needs, and show how an agent would run it.
