INDEPENDENT AI / API DEVELOPMENT

Small AI workflows.
Built to be verified.

I’m ShengRui Xu. I focus on Python APIs, structured LLM outputs, and practical automations—with a clear scope, visible tests, and a handover you can use.

Discuss a small project Explore the proof of work

Start with one problem. Agree on what “done” means.

A SMALL EXAMPLE. A CLEAR CONTRACT.

POST /inquiries

{
"Idempotency-Key": "demo-001",
"message": "Please send a quote."
}

201 created 1 inquiry / 1 audit event
200 replayed no additional write

validate inputcheck the keycommit together

Recorded local API scenario · not a live service
Headers and body abbreviated for display.

Python / FastAPIStructured outputsValidation & testingSmall-scope delivery

01 / SERVICES

A focused fix.
Not a full-platform promise.

Begin with a bounded task in your existing system. Agree on the input, deliverable, and acceptance check before implementation.

Python API fixes

Trace an invalid write, repair request validation, or improve a specific error path without rewriting your backend.

Deliverable: scoped patch + regression tests.

Structured LLM outputs

Define an output contract and handle malformed JSON, refusals, incomplete responses, and selected transient failures.

Deliverable: adapter + deterministic tests.

Small API automations

Connect one agreed input to an authorized API or downstream tool, with explicit state and failure handling.

Deliverable: narrow workflow + setup notes.

Availability and pricing are agreed after scope review. API usage, hosting, and ongoing support are not assumed to be included.

02 / SELECTED WORK

Inspect the behavior.
Not just the screenshot.

Includes a silent walkthrough, source code, and a test report.

INDEPENDENT DEMONSTRATION

API reliability &
validated model outputs

A local FastAPI + SQLite intake example, plus an independent LLM response adapter. Built with AI assistance and synthetic data; not a previous client project.

81tests passed · 2026-09-09

6recorded API scenarios

0live model calls in tests

View the case study

CASE 001 / API RELIABILITY

Local API and mock model output walkthrough, with labeled test outcomes

AI-assisted · synthetic data · local demonstration

03 / APPROACH

Small scope.
Visible evidence.

Understand the boundary

Clarify the business use, authorized code and data, expected behavior, and test environment.

Make the change

Reproduce the issue, agree on a small milestone, and make a focused implementation.

Verify and hand over

Show the relevant checks, explain the changes, and call out what remains unverified.

LET’S DEFINE THE FIRST MILESTONE

One clear problem
is a good place to start.

Share the workflow, a sanitized example, and what a successful result should look like. We can agree on scope before work begins.

Start on Upwork

View the brief template

Project goal:
Current stack / workflow:
Sanitized example of the problem:
Expected result / acceptance check:
Deadline and budget range:
Code/data access I am authorized to provide:
AI-assisted development allowed: yes / no / discuss

Upwork-sourced inquiries: please keep pre-contract communication and contracting on Upwork. This site does not accept payment or submit messages.

Small AI workflows.Built to be verified.

A focused fix.Not a full-platform promise.