Business Systems / Case study
AI operations agents for a Shopify brand
Hardening five AI agents that run day-to-day e-commerce operations
In short
A current, confidential engagement hardening five AI e-commerce operations agents for a Shopify brand: explicit limits, checks before consequential actions, and reviewable runs.
- State
- In progress
- Scope
- 5 operations agents
- Platform
- Shopify
- Client
- Confidential
- Built with
- Shopify / TypeScript
The brief
The problem
The brand already had AI agents doing real operational work across its store. They worked often enough to be useful, which is also what made them risky. An agent that is right most of the time still needs a clear boundary, a way to notice when it is wrong, and a person who can see what it did.
What we did
Our approach
We are hardening the agents the way we harden our own. Each agent's job and limits are made explicit. Actions that matter go through checks before they reach the store. Failures surface instead of passing silently, and every run leaves a record the team can review. The work is scoped agent by agent, so the store keeps operating throughout.
Features
What it does
- 01
Per-agent scope and limits
- 02
Pre-action checks
- 03
Visible failures
- 04
Reviewable run history
Results
What happened
- Five operations agents in scope, hardened one at a time while the store keeps running
- Explicit boundaries for what each agent may do
- Checks in front of consequential store actions
- Every run reviewable by the brand's team
- Engagement in progress; the client and outcomes stay confidential
Services
What this work shows we do
AI automation
Service pageModels wired into the tools your team already uses, with logs and limits.
Shopify development
Service pageStorefronts, themes and integrations for Shopify brands.
Business automation
Service pageRepetitive steps moved to code that runs the same way every time.
Related work
Built with the same services
Developer ToolsJarvisMCP
JarvisMCP gives AI agents one governed door into our company systems: CRM, search intelligence, infrastructure, memory and coordination. Agents find the capability they need, act through it, and leave a record. They never hold the keys.
Read the case study
Business SystemsButlerCRM
ButlerCRM keeps contacts, deals, email, calendar and tasks in one workspace that updates live for everyone in it. We rebuilt it on Elixir/Phoenix so changes appear for the whole team as they happen.
Read the case study
How we work
How a change reaches production
The same five gates for our own products and for yours. Nothing reaches your customers on the strength of a demo.
- 01 · Isolate
Its own branch and workspace
Every change starts in a separate copy of the code, so work in progress never touches what is live.
git worktree add .worktrees/your-change - 02 · Build
Engineers with AI agents
Agents write and test code beside us. They hold no passwords: every call they make goes through our gateway, which checks it and logs it.
agent → gateway → policy · audit · call - 03 · Prove
Evidence, then review
Tests, before-and-after screenshots of the running app, then independent reviewers. It does not move on until no serious finding is left.
review: 0 high · 0 medium → pass - 04 · Ship
Deployed and checked live
The pipeline builds and deploys. Then we check the running service itself, not just a green build.
replicas 1/1 · image sha matches · health 200 - 05 · Watch
Checked every night
Each integration we run is tested against its provider every night, so a changed API is caught before your team notices.
nightly: each integration vs its provider