Day job (Software Tech Lead, Oizom) / 2026

One stack per ticket: a dev pipeline where the ticket is the only interface

Each approved ticket gets a branch in every repo, its own services, a copy of the dev database and a preview URL. An AI agent works it from the ticket thread; a script merges, deploys to dev and rolls back.

Diagram: a ticket moves to Approved, gets its own branch, services, database copy and preview URL, the agent works from the thread, and Git Push runs a scripted merge, CI, deploy and health check with rollback.
Diagram: a ticket moves to Approved, gets its own branch, services, database copy and preview URL, the agent works from the thread, and Git Push runs a scripted merge, CI, deploy and health check with rollback.

Result

  • Per-ticket sandboxes about 10x cheaper and faster than with full databases
  • One dev VM at about ₹4–6k a month runs six tickets in parallel
  • Merge, deploy to dev, health check and rollback are scripted, not agent-driven

Stack

GoSQLite (dev drivers)DockerCI/CDClaude CodeVS Code in the browser

Built at my day job, where I lead the software team behind the Envizom platform. The code belongs to the company; this write-up comes from my public blog posts.

The problem

The platform is several repos: hardware drivers, a simulator, a backend, a frontend. A new developer needed days to get a working local setup. Once the team started using coding agents, a developer could keep three tasks in flight, and the mistakes moved out of the code and into the branches around it: a change on the wrong branch, a fix that only worked on one machine.

What I built

  • The ticket is the spec. A developer writes the task in plain English and drags it to Approved. No IDE, no terminal.
  • A full stack per ticket. The orchestrator provisions a branch in every repo, the ticket's own backend, drivers, simulator and frontend, a copy of the dev database and a private HTTPS preview URL.
  • The agent works from the thread. It posts the preview link when done; the developer tests it and replies in the same thread. Every prompt is saved with the work it produced, which doubles as the audit trail.
  • Shipping is a script. At Git Push, no agent is involved: commit, merge dev in, open and merge the PRs, wait for CI, roll new images onto the dev VM, run health checks, and roll back with a note in the thread if the VM is unhealthy. Production keeps a separate path where people decide.
  • Cheap copies. In the rebuilt backend every store has a dev driver; SQLite runs inside the Go process, so copying a database means copying a few files.

Result

Sandboxes about 10x cheaper and faster than with full databases. The dev VM that holds every developer's devbox and the ticket stacks costs about ₹4–6k a month and handles six tickets in parallel. A ticket can wait in QA for a week without blocking anything else.

Read more




You have requested the PDF version of this page!
This page was not designed or optimized for your request.


You have a redirct option for your expected request:



  1. Checkout the PDF version: https://firebasestorage.googleapis.com/v0/b/panth-personal.firebasestorage.app/o/portfolio%2FPanth-Patel-Resume.pdf?alt=media



  2. Back to website: https://panth.vardayinitech.in/