MAGENDOO / AI

Agentic engineering for commerce.

Build software with agents. Build agents for commerce.

We help commerce teams put AI to work in software development and operations — safely, inside your own infrastructure, repositories, engineering standards and security constraints. Marketing ships without waiting on tickets. Engineering ships without fearing releases. Your environment. Your code. Your controls.

20+ years building production commerce systems · Magento / Adobe Commerce · Shopify · PrestaShop · Odoo · Go · Python · MCP · Fixed-fee engagements
How we deliver

Grounded agents. Connected systems. A human interface.

An agent is only as good as what it knows about your platform — and what it's allowed to touch. Our delivery workflow rests on three practices. The tools we've open-sourced are working parts of it, published.

01
Ground
Agents work from your platform's actual documentation and your engineering standards — not from guesses. We built this grounding layer for Magento, Adobe Commerce, PrestaShop and Shopify for our own delivery work, then published it open source.
02
Connect
We connect the commerce backend through typed, permissioned tooling. Reads are scoped. Writes require confirmation. Every action is logged. Your infrastructure, your approval rules.
03
Deliver
Your team gets an interface, not a ticket queue. MageMCP shows the pattern in public: connect Claude or Codex to your store and ask — for the report, the catalog fix, the order status — instead of asking a developer for a module and waiting.
MageMCP: open source · 60 typed operations · 895 automated tests

The same workflow, applied inside your project and your constraints, is what we deliver.

The bottlenecks

The work that waits on a ticket.

Four patterns show up in almost every mid-market store we look at. None of them are staffing problems. They're workflow problems — and workflow is what this practice changes.

Marketing waits on developers.

The banner swap, the campaign page, the product copy fix — each one becomes a ticket, and every ticket joins a queue.

Reports wait on developers too.

Every "can you pull the numbers on…" lands on the one person who can query the store, somewhere behind the feature work.

Releases are events, not routine.

Deploys need maintenance windows, peak season means a code freeze, and the backlog keeps growing while everyone waits for a safe moment.

Integrations fail quietly.

Store, ERP and fulfilment drift apart in the background — and the first alert is a customer asking where their order is.

Agents don't fix these by magic. A disciplined workflow does — agents make it fast.

Definition

What is agentic engineering?

Agentic engineering is the discipline of putting AI agents to work on real software — planning, writing, testing and reviewing code under the same controls human engineers follow. Applied to commerce, it covers two things: using agents to build and maintain your store's software, and building agents that run parts of the operation itself.

We deliver it two ways: a fixed-fee Agentic Engineering Assessment and an ongoing Embedded Agentic Engineer engagement.

Why now

AI is entering commerce from two directions.

Buyers are arriving through AI. AI-referred orders on Shopify grew nearly 13x year-over-year in Q1 2026 and tripled again in Q2, and every major platform is shipping agent-facing infrastructure — ACP, UCP, MCP. Your catalogue, APIs and checkout were designed for people clicking pages. Increasingly they will serve software acting for people.

At the same time, agents are entering software development itself. Teams that establish disciplined agentic workflows ship faster with fewer defects. Teams that don't will be quoted against by teams that did.

Both shifts land on the same asset: your commerce engineering. That's what we work on.

Build with agents.

Turn your development process into an agentic engineering system. Within your existing controls, agents can:

implement features and modules
write and maintain tests
review pull requests
execute upgrades and migrations
refactor and document
investigate defects
prepare releases

Your developers stay in control. Delivery gets measured, not assumed.

Build agents.

Apply the same engineering discipline to commerce operations. Agents for:

catalogue and product enrichment
search and product discovery
merchandising
customer service
orders and ERP interactions
reporting and analytics

Built on typed tools, permissions, audit logs and write confirmations — not a chatbot bolted onto an API key.

How we work

Agents come to your environment. Not the other way around.

We don't ask you to hand your codebase, credentials or customer context to someone else's sandbox. We establish the agentic workflow inside your perimeter:

Your repositories
GitHub, GitLab, self-hosted
Your CI/CD and deployment process
agents open PRs; your pipeline decides
Your model provider
Anthropic, OpenAI, AWS Bedrock, Google Vertex, or your gateway
Your standards
PHPStan levels, coding standards, SonarQube, Snyk, Semgrep, coverage thresholds
Your approval rules
PR-only, required reviewers, protected branches
Your data boundaries
no source or customer data leaves infrastructure you control

If your rules say two approvals and no direct commits, those are the agent's rules too.

From first look to measured outcomes.

01
Assess
We inspect your architecture, code and workflow.
02
Deploy
We establish the agentic development environment inside your controls.
03
Embed
A senior engineer works with your team, not around it.
04
Measure
Delivery speed, quality, coverage, defects and human effort — against your baseline.
05
Improve
Automate progressively larger portions of the engineering lifecycle.

Two ways to work with us.

Agentic Engineering Assessment

Engineering assessment + commerce automation opportunity map.

Fixed fee, agreed before we start · typically 1–2 weeks

We go deep on how you build software: repository, architecture, engineering rules, CI/CD, testing, security, developer workflow — and design the private agentic workflow that fits your constraints. Alongside it, a survey-level opportunity map of where operational agents could create value across catalogue, search, service, orders and ERP.

You receive: an Agentic Engineering Scorecard, a workflow and architecture proposal, a prioritized backlog, a security and governance model, a 90-day implementation plan, and expected measurable outcomes.

Embedded Agentic Engineer

A senior commerce engineer inside your team, using agents to ship.

2 days/week · 3 days/week · dedicated

Not staff augmentation, not an agency project. Dedicated engineering capacity with defined objectives and monthly measurement — implementing the assessment roadmap, establishing agentic workflows your team keeps, and shipping production software along the way. Familiar with the Forward Deployed Engineer model? That's the model.

For organisations without internal AI leadership, the engagement can also cover architecture, governance, vendor selection, engineering standards and roadmap ownership.

The engineers describing the architecture are the engineers who ship it.

Magendoo is a senior-practitioner commerce engineering studio with 20+ years building production e-commerce systems — two decades of Magento and Adobe Commerce architecture for retailers and brands, alongside Go and Python platform work.

All of it now applied to one problem: making commerce engineering work with AI agents, safely. The open-source tooling on this site — MageMCP, the Magento modules, the agentic search middleware — is built and maintained by the same engineers who run your assessment and embed with your team.

No account managers. No hand-offs. Senior practitioner access, end to end.

Trusted with production commerce for two decades.

“Magendoo is really good — sharp and experienced, fluent in technical and business processes. They helped us move from a legacy system to a microservices architecture that handles meaningfully higher load at lower cost.”

Osvaldo Spadano, CTO, Akoova, UK

“It was a pleasure to work with Magendoo. Their proactive approach and constant drive for the right solutions make them a valuable asset on any team. They delivered our payment integration ahead of schedule and under budget.”

Rens Gerritsen, Technical Lead, Buckaroo, Netherlands

Field notes from the agentic frontier.

→ All insights
Top 20 Open-Source Search Engines and Product Discovery Tools for E-Commerce
Ranked list of 20 open-source search engines, vector databases, and product discovery tools for e-commerce, sorted by GitHub stars. Covers E…
June 25, 2026
Top 15 Open-Source PIM Systems on GitHub — Features, Stars, and E-Commerce Integrations
Ranked list of 15 open-source PIM systems and commerce frameworks with product management capabilities, sorted by GitHub stars. Includes Ake…
June 25, 2026
Forward Deployed Engineers in E-commerce
Executive summary TL;DR Bottom line The FDE role is already present in e-commerce, but usually under different names. Today’s commerce leade…
June 11, 2026

Find out what agents can do inside your engineering — safely.

Fixed fee, code-level, typically 1–2 weeks. You keep the scorecard, backlog and roadmap whatever you decide next.