Santekno
Projects/Shop Downgrade Experience
ByteDance logoPart of ByteDance · Software Engineer, Promotion Platform Team
Backend / APILive

Shop Downgrade Experience

Designed an end-to-end shop downgrade experience with layered architecture and centralized use case logic for reliable, reusable delivery.

2025-shop-downgrade-experience · main
# E-Commerce / Promotion
$ go -X POST /2025-shop-downgrade-experience/v1/sync
→ 201 Created
{ "status": "ok" }
publish 2025-shop-downgrade-experience.synced
Role
Backend Engineer
Timeline
2025
Team
2
Domain
E-Commerce / Promotion
Stack
7 tech
Status
Live
01

Overview

The Shop Downgrade Experience project defined how a shop transitions to a downgraded state within the promotion domain. I was responsible for designing the technical plan end-to-end — from the database schema and access layer through to the business layer — and for delivering it as a clean, reusable backend flow.

A backend has no screens, so its "UI" is the operational surface: the downgrade API, the layered flow, and unit-test coverage.

API contract — downgrade endpoints & fields
Ops view — release & monitoring flow
Test coverage — unit tests per task
03

Problem & Solution

Problem

Downgrade behaviour was needed across several scenarios, and implementing each one separately would have created redundant code and a heavier maintenance burden. The work also had to ship on a tight timeline without sacrificing quality or stability.

Solution

I designed a layered architecture and proposed centralized use case logic so a single core flow could be reused across scenarios instead of duplicated. I backed every development task with unit tests, then stayed involved through the release and monitoring phases to ensure a smooth, incident-free delivery.

04

Workflow

Three actors interact with the service; the diagrams are written in English to stay consistent across both languages.

flowchart diagram

downgrade trigger

resolve use case

access layer

metrics

Shop

Promotion Service

Central Use Case

RDS

Prometheus / Grafana

Use case interaction
05

Architecture

flowchart diagram

metrics

Seller / Shop

Promotion Service

Business Layer

Central Use Case

Access Layer

Managed RDS

Prometheus / Grafana

System architecture
Architecture notes

promotion-service receives the downgrade request and routes it through the business layer into a shared central use case, which reads and writes state through a dedicated access layer over a managed RDS (relational database). Keeping the use case centralized means new scenarios reuse the same validated path, while Prometheus/Grafana metrics cover the operational surface.

state diagram

trigger

use case applied

active

downgrading

downgraded

Shop downgrade lifecycle
06

Features

Design
  • End-to-end technical plan
  • Database schema design
  • Layered access & business design
Quality
  • Centralized use case logic for reuse
  • Unit tests per development task
Delivery
  • Release execution
  • Post-release monitoring
07

Tech Stack

language
Go
framework
RPCHTTP
database
Managed RDS
testing
Unit Tests
observability
PrometheusGrafana
08

Engineering Challenges

01
Reducing redundant development

Similar downgrade behaviour was needed across many scenarios. Fix: centralized use case logic so the same core flow could be reused instead of re-implemented, cutting effort and keeping the code maintainable.

02
Delivering quality without long timelines

A high-quality outcome was expected on a tight schedule. Fix: minimal yet strategic technical planning, unit tests for every task, and hands-on involvement through release and monitoring to keep the launch incident-free.

09

Outcomes

3
architecture layers designed
0
post-release incidents
100%
tasks covered by unit tests
10

Lessons Learned

A high-quality outcome does not always require lengthy development — centralized use case logic and disciplined unit testing let me streamline effort while protecting long-term maintainability. The project deepened my interest in taking on more architectural ownership.

11

Related Santekno Tutorials

Interested in working together?

Order, payment, or high-traffic event pipelines — let's talk architecture and trade-offs.