For small-business owners · Shanghai / Remote

Turn repetitive, people-dependent operations into
working AI tool systems

I map the real workflow first, then use AI automation, internal tools, or device integration to reduce repeated work and information gaps.

Discuss on WeChat See real cases

Customer problems

You may not need another AI tool. Your workflow may already be losing control.

Repeated work keeps growing

The same information moves between spreadsheets, chat, email, and business systems.

Work depends on a few people

Steps live in personal memory and break during handoff, absence, or growth.

AI is used but not dependable

Prompts stay with individuals, results are hard to check, and nothing enters the formal workflow.

Business outcomes

Start with a concrete result, not a technology label

Reduce repeated work

Make frequent research, drafting, synchronization, and reminder steps more dependable.

Make the process inspectable

Keep human approval, exception records, and accountability visible.

Create maintainable assets

Preserve rules, data, and operating notes so the business is not dependent on rescue work.

How it is solved

One objective, three implementation paths

Software, agents, hardware, and models are implementation tools. The right choice depends on workflow, data, and maintenance conditions.

AI Workflow Automation

Turn repetitive intake, research, drafting, handoff, and follow-up work into AI-assisted workflows with clear human checkpoints.

Common problems

The same information is copied across several tools
Key steps depend on one person's memory and are hard to hand over

Expected outcomes

Define which steps AI handles and where people remain accountable
Establish a repeatable workflow with visible exceptions

Internal Tools and Desktop Systems

Build software around the way the business actually operates when spreadsheets, scripts, and generic SaaS no longer hold together.

Common problems

Spreadsheet versions and permissions are becoming difficult to control
A script only works for its original author and is expensive to troubleshoot

Expected outcomes

Put core operations behind a clearer interface and data model
Reduce onboarding, troubleshooting, and handoff cost

Device, Data, and AI Integration

Connect existing devices to dependable data collection, control, teaching, operations, or AI-assisted workflows.

Common problems

Devices run, but their data never reaches the business workflow
Firmware, curriculum, and operating instructions drift apart

Expected outcomes

Standardize device communication, configuration, and telemetry
Establish reusable deployment, troubleshooting, and content-update practices

Case evidence

Inspect how the problem was handled before inspecting the technology

Client names may remain private, but my role, constraints, decisions, and result wording must stay explicit.

Case

Mobile Assistant for EdTech (Alias) →

Delivered a mobile assistant and Android learning tablet adaptation for education workflows, built with native Kotlin for robust UX and API integration.

Case

Global Brand Site Growth Revamp (Alias) →

Rebuilt information architecture and landing systems to improve distribution and conversion quality.

Case

Campus IoT Kit Upgrade (Alias) →

Redeveloped ESP32 + MicroPython education hardware kits with platform-linked teaching workflows, while contributing common libraries and examples to the Aily Blockly ecosystem.

View all cases→

Working method

Start with one real problem

01
Describe the current situation by WeChat or email; a polished requirements document is not required.
02
Decide whether the problem deserves development or can be solved with existing tools.
03
Validate one frequent and well-bounded step first.
04
Use real operating records to continue, adjust, or stop.

Why work with me

Eric Xie · AI Workflow and Tool Systems Engineer

My background spans enterprise software, EdTech R&D, and software-hardware projects. The working method is practical: first decide whether the problem deserves development, then choose automation, an internal tool, a desktop system, or device integration. I personally join the conversation, solution judgment, and key implementation instead of hiding business boundaries behind “universal AI.”