← All workshops

Build Your Software Factory With AI

Turn a work request into code your team can review. Use AI to connect software tasks to a repeatable build process. Start with a task in Linear, run the build in a separate environment, and test the changes. Send the code to GitHub for review. Build with prepared code and keep people in control of releases.

Your First Factory Run, Ready For Review.

  • Linear Work Orders

    A Linear work-order flow with acceptance criteria, run status, evidence, and draft-PR links.

  • Isolated Build Environments

    An isolated Vercel Sandbox build that investigates a bounded issue, implements a change, and runs checks.

  • Durable Orchestration

    A durable Workflow SDK pipeline with AI SDK tool orchestration, approval pauses, and bounded retry behavior.

  • Factory Blueprint & Runbook

    A tested ticket-to-draft-PR prototype, repository blueprint, regression cases, and an operator runbook.

Build The Pipeline. Prove The Handoff.

Day 1

From Ticket To Isolated Build

  1. Prepare The Repository

    Review the blueprint, repository instructions, acceptance criteria, and credential boundaries.

  2. Connect Linear And GitHub

    Build the work-order intake and useful status updates from sample or approved tickets.

  3. Investigate In Vercel Sandbox

    Run a small reproduction, inspect the repository, and prove the request before editing.

  4. Implement A Bounded Change

    Use AI SDK tools and durable Workflow SDK steps to produce a candidate change and saved run evidence.

Day 2

Verify, Recover, And Hand Off

  1. Verify The Actual Change

    Run tests and inspect the real diff from a fresh verification context.

  2. Exercise Failure Paths

    Test retries, duplicate events, missing context, approval pauses, and cost limits.

  3. Prepare The Draft PR

    Publish a reviewable change in a repository you control and link evidence back to Linear.

  4. Review The Factory And Next Rollout

    Demo one ticket-to-PR run, identify remaining work, and schedule your private implementation review.

Files You Can Put To Work.

Adapt these blueprints to a small repository, run the supplied cases, and document what your team needs before a wider rollout.

  • Factory starter codebase

    A prepared TypeScript ticket-to-draft-PR blueprint with adapters for the guided stack.

  • AGENTS.md + skills

    Repository context, investigation, implementation, verification, and Linear-update instructions.

  • Work-order schema

    Ticket, acceptance criteria, state, approval, cost, and evidence fields for each run.

  • Regression case pack

    Supported change, ambiguous ticket, false premise, build failure, and duplicate-event cases.

  • Operator runbook

    Credential scope, budgets, review gates, failure recovery, cleanup, and rollout checklist.

Facilitators

Meet Your Facilitators

Santos Hernandez

Santos Hernandez

Founder Ship AI, ggbucks, & Cofounder desic

Is This Your Next Step?

  • A good fit for

    Developers, technical founders, engineering leads, and small product teams that already ship software with coding agents and want a repeatable delivery system.

  • What to bring

    Bring a laptop, GitHub and Linear accounts, a Vercel account, model API access, and a small TypeScript repository you can safely experiment on. You should be comfortable with Git, pull requests, Node.js, and basic tests. Use our starter repository if your company code cannot be shared.

  • Included

    A blueprint to build from, expert guidance while you build, and recordings and materials emailed afterward.

  • A review after the lab

    A 30-minute private readiness review before the lab and two private implementation-review hours within 30 days afterward. We review one repository and your workflow evidence; this does not include done-for-you rollout or unlimited support.

Tools For Each Stage Of Delivery.

Linear tracks the work; Sandbox provides an isolated execution environment; Workflow SDK coordinates durable steps; AI SDK connects model calls and tools. GitHub holds the code and draft PR. This is an independent Ship AI workshop.

A Few Things To Know.

What Preparation And Follow-Up Are Included?

A 30-minute private readiness review before the lab and two private implementation-review hours within 30 days afterward are included. Use them to scope one repository, review run evidence, and plan your next rollout steps. This does not include unlimited support or done-for-you implementation.

Is This Different From Build And Test Software With AI?

Yes. Build And Test Software With AI teaches the individual build workflow: worktrees, tests, browser checks, and skills. This advanced workshop connects ticket intake, durable orchestration, isolated execution, verification, and progress updates into a software-delivery system.

How Is This Different From Test And Improve Your AI Systems?

That workshop improves the reliability and operation of an existing AI workflow. Software Factory applies those ideas specifically to software delivery: turning a ticket into repository evidence, a tested change, a draft PR, and a Linear update.

Will The Agents Automatically Merge Or Deploy My Code?

The guided build stops at a tested draft PR and reviewed status update. Human approval controls merge and production release. We practice one bounded repository workflow; a complete autonomous engineering organization is outside the two-day scope.

Is This An Official Linear Or Vercel Event?

No. This is an independent Ship AI workshop using those tools. It is not a vendor certification or a claim of sponsorship.

Which Tools And Accounts Do I Need?

The guided stack uses Linear for tickets and run updates; GitHub for code and draft PRs; Vercel Sandbox for isolated execution; Workflow SDK for durable steps; and Vercel AI SDK for model calls and tools. Supported connections depend on account plans and administrator approval. Sample tickets and a starter repository are available; the final setup guide will specify tested versions.

Are Software Subscriptions Included?

Admission does not include Linear or Vercel plans, model API usage, Sandbox compute, hosting, or other software costs. We start with a small bounded task and configure run budgets. Bring accounts you control; never share raw credentials in the workshop.

When And Where Will It Take Place?

This is a two-day lab capped at 12 attendees. The date, daily schedule, and venue will be announced before paid registration opens.

What Is The Refund Policy?

Once paid registration opens, tickets will be refundable until the day of the event. If the event is cancelled, you will receive a full refund. Email hi@shipai.club with questions.

Something else on your mind? Ask us.

From First Agent To Software Factory

Start with a useful agent, move into code-based workflows, then develop the engineering and reliability skills for a delivery system.

  1. $299Build Your First AI Agent
  2. $499Build Your First AI Workflow
  3. $999Build And Test Software With AI
  4. $2,999Test And Improve Your AI Systems
  5. You are hereBuild Your Software Factory With AI

Not sure which one? Take the one-minute assessment