Ningbo · Apps / Software / AI Agents

Good ideas,
built into real products

Fengqi Studio is a small product team. We build mobile apps, custom software, and AI agents that actually get work done — from first discovery to launch and long-term support.

8+ yrs
Product experience
120+
Projects delivered
30+
Long-term clients

Industries we serve

Fintech Healthcare Retail Logistics Education Manufacturing

Our Services

Four tracks across the whole product lifecycle

From an early idea to long-term operation, you don't need separate teams for product, engineering, AI and maintenance. We offer one continuous track — or plug into the team you already have.

Mobile Apps

iOS · Android · Cross-platform

End-to-end delivery for iOS, Android and cross-platform apps — from interaction design and development to real-device testing and store submission.

  • iOS
  • Android
  • Flutter
  • App store
Learn more

Custom Software

Web · SaaS · Mini programs

Business systems, SaaS platforms, mini programs and admin consoles — designed around your real workflows, not forced into a template.

  • Web systems
  • SaaS
  • Mini programs
  • Admin
Learn more

AI Agents

LLM · RAG · Automation

Connect large models to your business: knowledge-base Q&A, automated workflows and multi-agent collaboration with measurable outcomes.

  • LLM
  • RAG
  • Workflows
  • Private cloud
Learn more

Advisory & Care

Review · Optimization · Support

Architecture review, performance tuning, technology selection and long-term maintenance — for teams that only need the missing piece.

  • Architecture
  • Performance
  • Tech selection
  • Support
Learn more

AI Agent Development

Not just a chatbot —
an agent that delivers results

We connect models to real business operations: retrieving company knowledge, handling tickets, generating reports, and coordinating multiple roles. Controlled processes, grounded answers, measurable impact.

Explore AI capabilities
  • 01
    Knowledge-base Q&ARAG-based answers with citations
    Knowledge
  • 02
    Automated workflowsSupport, tickets and reports
    Automation
  • 03
    Multi-agent collaborationRoles split complex tasks
    Multi-Agent
  • 04
    Private deploymentData stays inside your network
    Security
Project HealthOn Track
98% On-time delivery · 2024–2026
Q1Q2Q3Q4

Why Fengqi

Why a small team is often more reliable

01

Talk directly to the people who build

No account managers relaying messages. What you say in a call shows up in the next build.

02

Progress you can actually click

Two-week iterations. Every round ends with a working version you can open and test.

03

Scope gets trimmed first

Start with the 20% that matters most, validate cheaply, then decide whether to scale up.

04

Launch is not the finish line

We stay around after delivery: watch the data, fix issues, and keep iterating.

How We Work

Five stages, each with clear outcomes

No project by vibe. Every stage has defined outputs, reviewable results and a decision point for the next step.

STEP 01

Discovery

Clarify the goal, users and boundaries. Decide whether it is worth building — and whether it can be built.

STEP 02

Proposal & quote

Prototype, technical route and a clear timeline. One transparent price, no vague estimates.

STEP 03

Design & build

Two-week iterations with design and engineering moving in parallel. Progress is always visible.

STEP 04

Test & launch

Functional testing, load testing and real-device acceptance before deployment.

STEP 05

Operate & iterate

After launch we watch the data, collect feedback and improve by priority.

Selected Work

Things we have built

All work
CloudNest Wallet app interface Fintech

CloudNest Wallet

Mobile finance app

A mobile wealth app for younger users, combining accounts, an asset overview and smart bookkeeping into interfaces that read at a glance.

iOS / AndroidAsset management12 weeks
Answerly AI agent interface AI agent

Answerly AI

Knowledge-base agent

An internal support agent connected to 20,000 documents. Answers cite their sources; complex cases are handed to a human automatically.

RAGKnowledge basePrivate deployment
GleamRetail store dashboard Retail

GleamRetail

Retail SaaS platform

A lightweight SaaS for chain stores. Inventory, members, orders and dashboards in one place — managers know what to restock each morning.

SaaSInventoryDashboards

Clients Say

What clients say

From brief to launch took ten weeks, with zero delays. Every weekly demo was something we could open and try — that gave us real confidence.

W
Ms. WangCloudNest Finance · Product Lead

They understand the business, not just the backlog. Several things we hadn't figured out ourselves, they raised first.

C
Mr. ChenGleamRetail · Founder

After the AI support agent went live, manual tickets dropped by about half, and answers became more consistent. It runs on our own servers.

L
Mr. LiCarePath Medical · Operations

Tech Stack

Proven technology, stable products

Swift Kotlin Flutter React Native TypeScript React Vue Node.js Python Go PostgreSQL Redis Docker Alibaba Cloud Tencent Cloud LangChain Dify RAG Vector DB

Start a Project

Let's talk about your next product

Whether it is a rough idea or a rebuild, start by telling us what you have in mind. The first conversation is free, and we reply within 48 hours.