*AI routing depends on configuration; Workers AI also provides document conversion and web search. Browser Run handles Gadget exports and can be omitted. †Basin Pipelines (formerly Pipelines): optional PRODUCT_ANALYTICS binding; absent from the default release configuration. Audited Oct 22, 2026 · public source a679442 (Oct 20) · Bindings · AI routing · Analytics hook · Product names
FOUNDATION
Built on Workers, by the Workers team.
WORKSPACE DURABLE OBJECT
One OverseerDurableObject per workspace
Access checks · orchestration strongly consistent workspace state
SUPERVISOR
Workspace supervisor
Controls capabilities, orchestrates agents and manages Gadget and Gatekeeper facets.
Own SQLite database
WORKER LOADERLoads the Gadget's exported class
GADGET FACET
AI-built application
Dynamic Worker code; no direct outbound network access.
Isolated SQLite
INTEGRATION WORKERSupplies a Gatekeeper class
GATEKEEPER FACET
Governed connection
Scopes remote-service access and mediates actions.
Agents are a different workload than websites — they chain tool calls, wait on LLMs, and need to run close to whatever they're calling.
DESIGN NEED
WHAT YOUR WORKFLOW NEEDS
CLOUDFLARE FOUNDATION
Execution
Run generated app code in isolation
Workers + Dynamic Workers
State
Coordinate shared, interactive work
Durable Objects + isolated facets
AI traffic
Choose models and govern requests
AI Gateway + model providers
Access
Connect approved systems and users
Gatekeepers + configured Zero Trust
Complement your existing cloud investments. Deep managed ML training and GPU pipelines remain hyperscaler strengths; this foundation focuses on serving and securing connected AI workflows.
ARCHITECTURE, NOT ASSEMBLY
Purpose-built, not assembled from parts.
On traditional cloud, an agent means stitching together six services before writing any agent logic.
Compute service
Database
Message queue
Container runtime
Secrets manager
API gateway
On Cloudflare, it's one integrated framework:
Agents SDK
Durable Objects — state lives with the agent
Workflows — durable retries + human-in-the-loop
THE PRICING ARGUMENT
Pay for thinking, not for waiting.
Agents spend most of their time waiting — on the LLM, on a human approval, on a scheduled trigger.
Traditional VM / container alive 24 hours
24 hours
Cloudflare Durable Object alive 24 hours, hibernates when idle
10 seconds
Workload illustration, not a price quote: CPU billing tracks compute, while Durable Object duration depends on activity and hibernation. Requests, storage, models and other service charges also apply.
THE NEW PRIMITIVE · CODE MODE
Don't call tools — write code.
Models are trained to write code. So give the agent one code tool, not fifty — it writes a plan that runs in a sandbox and returns only what matters.
DIRECT TOOL CALLS
Every step round-trips the model
LLM→tool→LLM→tool→LLM→tool
CODE MODE
One program composes it all in-sandbox
LLM→code · many calls · one result
fewer hopsless repeated model orchestration
// the model writes this — not 5 separate tool calls const projects = await codemode.list_projects({ status: "active" }); const blocked = []; for (const p of projects)
blocked.push(...await codemode.list_tasks({ projectId: p.id })); return blocked.filter(t => t.status === "blocked");
A Developer Platform primitive — code runs in an isolated Dynamic Worker, so credentials stay out of the model context. On MCP server portals, Code Mode is on by default.
05
From AI potential to business value.
NTT DATA connects the platform to your people, systems and operations.
WHAT NTT DATA BRINGS
Your transformation. Expertise at every step.
NTT DATA-led engagement
Discover the right opportunity
Assess AI readiness
Prioritize workflows
Define business outcomes
Connect your enterprise
Identity and access
Selected SaaS and data
Custom integrations
Build around your teams
Curated knowledge and skills
Tailored apps
Repeatable workflows
Govern AI adoption
Scope resource access
Configure approvals
Review models and usage
Operate and improve
Support user adoption
Maintain integrations
Review and evolve outcomes
YOUR KNOWLEDGE, MADE ACTIONABLE
Turn your expertise into reusable capability.
Business Knowledge
Your approved sources and guidance
Expert Workflows
How your best teams get work done
↓
NTT DATA + Your Experts
Select, structure and validate
↓
Shared Context + Skills
Reusable collections with scoped sharing
↓
AI Agent Workspace
Surfaced and ready-to-use in Cloudflare OS
NTT DATA helps turn institutional knowledge into a maintained capability for your teams.
1
Select trusted knowledge
Identify owners, approved sources and access boundaries.
2
Capture expert practice
Document the steps, decisions and exceptions that matter.
3
Package reusable skills
Give agents task instructions and relevant business context.
4
Validate with your teams
Check output quality, permissions and human review.
5
Maintain and improve
Keep knowledge current as your systems and processes change.
KNOWLEDGE ENGINEERING · NTT DATA
Two file types, one knowledge layer.
Skill File
Your repeatable way of working — expressed as task instructions.
Example: prepare a weekly account review
Steps, checks and escalation conditions
Reusable across the teams you authorize
One page ideal · .md
Context File
Your business knowledge — available where the task needs it.
Example: products, terminology and service guidance
Approved reference material and business rules
Owned and maintained with your subject experts
One–two pages · .md
YOUR DELIVERY JOURNEY
Start focused. Scale with evidence.
NTT DATA leads a phased engagement, with your business owners involved at every decision.
01
Discover
ALIGN
Select one valuable workflow
Assess data, access and readiness
Agree scope and success measures
02
Pilot
PROVE
Deploy a scoped workspace
Connect approved resources
Configure skills and controls
Measure results with pilot users
03
Expand
EVOLVE
Plan rollout from pilot evidence
Agree operational ownership
Improve context and workflows
Agree scope, integration feasibility, timeline and support terms during discovery. Expansion follows readiness and measured results.
YOUR PILOT SCORECARD
Measure what matters to your business.
Establish the baseline first. NTT DATA works with your team to assess the change.
Workflow efficiency
Completion time, manual steps and human review effort per task.
Quality and adoption
Output acceptance, corrections, repeat usage and user feedback.
Cost per task
Model and platform usage, integration and service costs against delivered value.
Governance fit
Resource scope, approvals, access reviews and operating ownership.
Evidence before expansion. Targets are agreed for your workflow during discovery. Productivity gains and savings are measured, not assumed.
06
Your adoption journey.
One meaningful workflow. A clear owner. A measured next step.
START WITH NTT DATA
Bring us one workflow.
1
A business problem worth solving
A repeated task with a clear owner and a baseline you can measure.
2
The people and systems involved
Business experts, IT and security — plus the resources the workflow actually needs.
3
A shared definition of success
Agree the pilot scope, controls, evaluation criteria and next decision.
YOUR OPERATING MODEL
Your teams stay in control. NTT DATA helps you deliver.
BUSINESS
Own the outcomes
Your business owners select priorities, validate outputs and guide adoption.
TECHNOLOGY
Shape the deployment
NTT DATA and your IT and security teams align identity, integrations, data access and models.
OPERATIONS
Keep improving
Agree support responsibilities, release management and ongoing knowledge and workflow maintenance.
Build on existing App Security and Cloudflare One investments where they fit your architecture.
NEXT STEP
Let's build your AI-powered workplace.
NTT DATA helps you turn AI experimentation into secure, connected business workflows — powered by Cloudflare OS.
MOVE 01
Discover your first use case
Bring one repeated task, a business owner and the systems involved. We will shape a focused pilot together.
MOVE 02
Prove the outcome
Agree success measures, validate access and outputs, then use the evidence to plan your next step.