Open every service repository under one parent directory so Claude Code can trace a customer question across the whole product. Combine current code with product documentation and a customer quirks page to produce specific technical answers and deployment plans.
Source episode · 04:50
How Al Chen Uses Claude Code and 15 Repos to Answer Any Customer Questionwith Claire Vo
Al shows fifteen service repositories opened under one VS Code workspace, then combines current code, Confluence documentation, and customer-specific deployment constraints in Claude Code.
A customer-specific technical answer or deployment plan grounded in current code, documentation, and known environment constraints.
Claude for coding tasks
Step by step
Follow the sequence once, then adapt the prompts, checks, and handoffs to your own setup.
5 steps
Clone all of your company's individual code repositories into a single parent directory on your local machine. This setup allows an AI assistant to see the entire codebase at once, rather than being limited to a single service.
In VS Code, open the top-level parent directory that contains all the individual repositories. This is a key step that enables your AI assistant to search and understand connections across every service in the platform.
Create one command that updates the expected main branch in every repository under the parent directory. Make it stop or report clearly when a repository has local changes, a missing remote, or a failed pull.
Create a script named pull-all that finds each Git repository directly under this directory and updates its main branch from origin. Do not discard local changes. For every repository, print the branch, whether it updated, and any error that needs attention. Exit nonzero if any repository could not be updated.
Save the generated script and run it daily to keep your local codebase in sync with the main branches.
Maintain one lightweight customer quirks page with the deployment, security, identity, networking, and data requirements that make each enterprise environment different.
I am creating a 'Customer Quirks' document to track unique requirements for our enterprise customers. Generate a template with sections for [Customer Name]. The template should include placeholders for key technical details like their security protocols, infrastructure specifics (e.g., secrets management, namespaces), and any environmental constraints we need to respect during deployment.
Ask Claude Code to answer the customer question using the current repositories first, then supporting documentation and the matching customer notes. Require code and document pointers so the result can be checked quickly.
Customer: [name] Question: [exact technical question] Desired result: [answer, deployment plan, or troubleshooting steps] Use the current code in this workspace, our connected documentation, and the customer quirks page. Account for these known constraints: [constraints]. Return: 1. A direct answer 2. Step-by-step implementation or troubleshooting guidance 3. The relevant repositories, file paths, symbols, and documentation pages 4. Any assumptions, conflicts, or facts you could not confirm Do not invent behavior that is not supported by the code or documentation.
Mentioning the customer's name in the prompt helps the AI cross-reference your 'customer quirks' page for a more tailored response.
Turn an idea into a PRD, user stories, and a plan.
After the steps
How to recover when the loop fails and where human judgment helps.
Recover
Keep building
Use a hybrid AI workflow to bridge design and development. Generate high-level design concepts and tokens with Claude, then use Codex for robust implementation of the technical components in your codebase.
Implement a QA loop for AI-generated code using Devin. This playbook combines automated code reviews with an agent-recorded video walkthrough for visual verification, ensuring high-quality merges at scale.
Build a reusable 'Watchdog' playbook in Devin to automatically monitor customer accounts. The agent proactively checks for errors, synthesizes a prioritized list of critical issues, and saves you time.
Join 100,000+ product managers who use ChatPRD to write better docs, align teams faster, and build products users love.