A dedicated development team for your system, by the month.

VAON provides a dedicated monthly development team of a BrSE, a Tech Lead and engineers. You talk to the engineers directly or through a Japanese-speaking BrSE, and AI-written code is always read by a person and tested.

A stable team works only on your project, and AI helps it move faster. No clear spec yet? Start with a 1 to 2 week Spec Lock.

What problems are you facing?

IT department is overloaded, leading to delayed deadlines

Existing vendors only code to instructions, nobody proposes a better way

Inconsistent deliverable quality

VAON's Team Model

Each team covers every role from requirements to deployment and operations. Headcount and roles scale with the project.

VAON team model diagram: a management and requirements layer (PM/BrSE, business analyst, UI/UX Designer), a technical leadership layer (Tech Lead/Architect) and a delivery layer (Developers, QA/QC, DevOps, Operations and Maintenance).

Illustration

Traditional multi-layer model vs.
VAON POD Model

Traditional multi-layer model

VAON POD Model

Communication through multiple intermediaries (Sales → PM → Translator → Dev)

You talk to the engineers directly, or through a Japanese-speaking BrSE. No layers in between.

Vendors only code as instructed, no one proactively proposes solutions

Involved from upstream: specs, wireframes and architecture.

Limited testing processes, often resulting in bugs on production

A feature counts as done only when it has tests, has been reviewed by a person, has passed security gates and you have accepted it.

Hidden costs and high dependency on specific individuals

AI-written code is always read by a person and tested. You can see the review and test records.

Lack of visibility into actual progress and quality

Weekly report: progress, risks, decisions we need from you.

VAON's Difference

Signed off

AI-written code

Always read by a person and tested. You can see the review and test records.

Fixed price

To start: the Spec Lock

1 to 2 weeks: requirements, wireframes and data design at a fixed price quoted upfront. You can stop afterwards.

Direct

With engineers, or a Japanese-speaking BrSE

Talk straight to the people doing the work. Every agreement is put in writing.

Technical Capabilities

Backend

Java
Node.js Python Django Laravel

Frontend

React Vue Next.js TypeScript

Mobile

Flutter Swift Kotlin React Native

Database & Search

PostgreSQL MySQL Redis OpenSearch

Cloud & DevOps

AWS Google Cloud Docker Terraform

E-Commerce

Shopify EC-CUBE

AI & Data

Python OpenAI LangChain Pinecone

OCR, LLM integration, RAG (answers grounded in your documents)

Workflow

  1. Spec Lock

    1 to 2 weeks to clarify requirements, wireframes, architecture and priorities.

  2. Team kickoff

    Agree on the point of contact, meeting rhythm, reporting and what "done" means.

  3. 1 to 2 week cycles

    Plan, build, and demo working software at the end of each cycle. Scope changes are quoted in writing.

  4. Quality control

    Three reviews, tests with every change, security gates on each pull request, a security check before every release.

  5. Weekly report

    Progress, risks, decisions we need from you.

Done means: tests · human review · security gates · your acceptance.

Quality Commitments

Our commitments are built on transparency, disciplined execution and continuous improvement.

1

Transparency

Weekly reports; review and test records you can see.

2

Quality

Bug severity levels and how each is handled are agreed at the start.

3

Stable Delivery

1 to 2 week cycles and planned releases; scope changes go through a written process.

4

Security

NDA, role-based access, code and data go into AI tools only with your permission, access revoked at the end.

5

Continuous Improvement

At the end of each cycle, the team reviews what slowed down or slipped through and picks 1 to 3 improvements.

Not sure how big a team you need yet?

If your spec isn't clear yet, start with a 1 to 2 week fixed-price Spec Lock. You get requirements, wireframes, architecture and a recommended team size.

Learn about the Spec Lock Phase

Choose the right engagement model

Dedicated monthly team (Labo)

Recommended

A stable long-term team on your system
Priorities adjusted at the start of each cycle
Direct contact with engineers or through a Japanese-speaking BrSE
The quote lists every role (BrSE, Tech Lead, engineers) and person-months, with nothing outside it

Best for: projects of 6 months or more.
Payment: per person per month, within an agreed budget cap

Book a free call

Project-based, fixed price

Scope and budget set upfront
Paid by delivery milestone

Best for: projects under 6 months with clear scope.
Payment: fixed price per milestone

Book a free call

Every project gets its own quote after the first call. Ownership of source code and IP is set out in the contract.

FAQ

The core team is ready in about 1 to 2 weeks. If you need more people, let us know 1 to 2 weeks ahead.

A Japanese-speaking BrSE acts as your point of contact when you need one. You can also talk to the engineers directly. Every agreement is put in writing.

Project knowledge lives in the team and the documents, which record the reasons behind technical decisions, and our CTO and a Tech Lead are on every project. If someone has to change, we give one month's notice and hand over within the team.

Our team works 8:00 to 17:00 Vietnam time, which is 10:00 to 19:00 in Japan, so our hours almost match yours.

Before work starts, VAON signs an NDA and agrees with you which data may be used and who may see it. Access is granted by role, and your code and data go into AI tools only with your permission. Every change goes through secret scanning, dependency scanning and static analysis. When the project ends, VAON removes its access.

Backend: Java, Node.js, Python, Django, Laravel. Frontend: React, Vue.js, Next.js, TypeScript. Mobile: Flutter, Swift, Kotlin, React Native. Database and search: PostgreSQL, MySQL, Redis, OpenSearch. Cloud and DevOps: AWS, Google Cloud, Docker, Terraform. E-commerce: Shopify, EC-CUBE. AI and data: OCR, LLM integration, RAG.

VAON engineers use Claude Code and GitHub Copilot every day, but no code is merged just because AI wrote it. The author re-reads it, AI reviews it, another engineer approves it and the team lead decides the merge. Review and test records are kept for every project, and you can see them.

Another question? Email us at sales@ai.vaon.com.vn

Even if you don't know how big a team you need yet.

The first 60-minute call includes our CEO, needs no documents and comes with no commitment.

Book a free call