SCOPE
Enterprise AI Agent Delivery · Rev. 2026.10
COMPANY
YuanASI · Shenzhen
CAPABILITY
Consulting / Development / Integration

Enterprise AI delivery · advisory, development & integration

Put AI agents into
real business workflows.

For companies that already have a defined, high-frequency workflow and need AI wired into existing systems: YuanASI handles assessment, an optional POC, build and integration, production deployment, and team handover. Our foundation is open-multi-agent, the open-source framework we build and maintain.

DELIVERY PATH · FROM BUSINESS PROBLEM TO PRODUCTION HANDOFFDWG. DELIVERY-01
  1. 01Assess
  2. 02Validate POC (optional)
  3. 03Build & integrate
  4. 04Deploy
  5. 05Handover
01
Case libraryClient projects in 4 industries
24Cases
02
Our own framework · open-multi-agentBuilt and maintained by YuanASI; third-party projects build on it
6,975GitHub Stars
03 / TECHNOLOGYDWG. OMA-SERIES · SHEET 03 / 05
OPEN SOURCE RUNTIMEopen-multi-agent

Control the plan.
Prove every run.

Open Multi-Agent (OMA) is the open-source framework YuanASI builds and maintains, and the foundation of our delivery work. It generates and runs task DAGs inside TypeScript / Node.js applications, with plan and per-task approvals, execution receipts, traces and a Run Viewer, EvalSets, and task-level checkpoints. Your organization chooses and controls the models, tools, credentials, and deployment.

TASK-LEVEL CHECKPOINTLOCAL / OFFLINE / AIR-GAPPEDPROCESS / ACP AGENTS
RELEASE SNAPSHOT · v1.13.0

Replay of a real run

↗
The official Run Viewer exposes the task DAG and span waterfall, including status, assignee, tokens, tool calls, and the full trace.
01CONTROL

Plans you can govern

Preview and approve a plan or an individual dispatch, require roles and ordering, and gate every default-deny tool call.

02EVIDENCE

Runs you can inspect

Stable run IDs, execution receipts, and traces feed a local Run Viewer or export to OpenTelemetry.

03EVALUATION

Quality you can measure

Versioned EvalSets, scorers, offline reports, CI gates, and online sampling support regression decisions before and after release.

04 / SERVICESSHEET 04 / 05

Enterprise AI agent systems
built to keep running—
with measurable business value.

Each service covers a stage of the delivery path; start from whichever stage your organization is in today. Every service has defined inputs, engineering outputs, and acceptance evidence.

SERVICE01
DWG D-01
VALIDATION GATEBUSINESS SCENARIOEVALSET + BASELINEPOCGO / NO-GO

Enterprise AI Advisory

Prove the case before expanding the investment.

  • Business outcomes, data boundaries, and non-AI constraints
  • POC plus a versioned EvalSet and reference cases
  • Acceptance report and Go / No-Go recommendation
Explore Enterprise AI Advisory →
SERVICE02
DWG D-02
OWNERSHIP TRANSFERSOURCE + TESTSDEPLOYMENTRUNBOOK + SOPYOUR TEAMBUILD · OPERATE · TAKE OVER

Custom AI Agent Development

Runs in your environment; your team takes over after delivery.

  • Node.js / TypeScript engineering assets and private deployment
  • OpenTelemetry integration, runbooks, and troubleshooting paths
  • Custom project source, tests, SOPs, and team training
Explore Custom AI Agent Development →
SERVICE03
DWG D-03
CONTROLLED EXECUTIONINPUTPLANAPPROVAL GATETASK ATASK BTRACE

Multi-Agent System Integration

Connect agents to existing systems and encode execution boundaries.

  • Dynamic or explicit workflows plus enterprise APIs and data
  • Plan approval, tool policy, timeouts, budgets, and task checkpoints
  • Trace / Run Viewer, regression evaluation, and CI gates
Explore Multi-Agent System Integration →
INTAKE / NEXT STEP

After we review your brief, projects that fit get a 30-minute scoping call.

→ Submit a project briefNo project scoped yet? Get an AI readiness report first
05 / INTAKE · SHEET 05 / 05

Ready to ship AI agents?
Start with a project brief.

Email

co@yuanasi.com

Business inquiries

Office

Nanshan District, Shenzhen

Keyuan West, Building 23 · North 2F-276

Company

Shenzhen YuanASI Technology Co., Ltd.

Founded 2023 · Shenzhen