WATER Agentic is coming soon.

WATER Agentic Platform

Coming soon

WATER Agentic Platform

The Platform is the runtime every WATER agent is built on: one conversation engine for every channel, shared behaviour joined to agents from settings, a model gateway, agent memory and a work engine that puts humans and agents on the same cases.

Where agents run.

What it does

Run agents

Available
  • One agent runtime . Every channel (web, WhatsApp, SMS, email, Slack) runs the same agent turn, so agents stay channel agnostic.Available
  • Agent SDK . Build an agent from one base class; tools are reached by skill through the gateway, and an agent can find the tools it needs.Available

Share behaviour without code

Available
  • Aspects . Memory, guardrails, the clock and confirmation are declared in an agent's settings and woven in by the runtime. They are never wired into agent code, so every agent behaves the same way.Available
  • Product and workspace layers . Base settings, then product settings, then workspace settings. Each layer can only add what it is allowed to.Available

Use any model

Available
  • Model gateway . Models are chosen by role (chat, judge, embeddings, rerank, extraction, vision) with retries and a circuit breaker. Every agent has its own key and spend limit.Available
  • Hosted or local models . Run hosted models or local open models behind the same gateway.Available

Remember safely

Available
  • Three memory tiers . Working memory for the conversation, episodic memory for what happened, semantic memory for what is known.Available
  • Policy before every write . Personal data is masked, and a policy decides what may be remembered.Available
  • Retention and purge . Memory expires on schedule and can be erased on request.Available

Put humans and agents on the same work

Available
  • Work engine . Cases, human and agent tasks, artefacts, timelines and SLA deadlines.Available
  • Work console . Staff inbox, queues, task review, case view and process board: one app for every solution.Available

See what agents do

Available
  • Tracing . Each turn's model, tools, cost and time.Available
  • Observability dashboards .Roadmap

How it works.

01

Channel

02

Runtime (aspects woven: guardrails, memory, clock, confirm)

03

Model gateway and tools via the gateway

04

Work engine

05

Work console.

Who it is for

CTO and head of AI

AI platform team

Solution architects.

Works with the rest of the suite.

WATER Agentic Tools

Agents that act.

  • Ready tool servers on the open MCP standard
  • Swap the system behind a tool by settings
  • Every call checked before it runs
Learn more

WATER Agentic Governance

Prove every decision.

  • Default deny on every tool, model and memory call
  • Guardrails, personal data masking, kill switch
  • Decision records for audit and risk
Learn more

WATER Agentic Industry Solutions

Agents for your domain.

  • HR operations first, insurance next
  • Shared domain knowledge reused by every agent
  • Connects to your systems of record
Learn more

Frequently asked questions.

Do I have to use your models?

No. Models are chosen by role in settings.

Can I add my own agent?

Yes. Agents are built with the SDK and registered in Governance.

Does it run in my cloud?

It runs in containers; deployment options for customers' own environments are on the roadmap. ---

COMING SOON

Governed agents are coming.
Get in early.