How PMs can start coding with AI: one problem, one workflow, one shipped result
Pick one real use case, choose one AI coding tool, and ship a first usable result in 48 hours.
Focus: tool choice · first ship · confidence
Pick a path and jump back to the interactive view
Each path opens short context + subscribe inside the interactive story.
Pick the path that matches what you want right now.
As AI coding gets cheaper and faster, durable advantage shifts from raw implementation to product judgment, communication, UX taste, and distribution. This site shows how builders and PMs can turn AI into shipped product, not just prototypes.
Learn the exact phone-to-Mac workflow: Codex on Mac, SwiftUI client, Xcode build pipeline, Tailscale networking, and OTA installation back to iPhone.
OpenA beginner-friendly path for product managers, operators, and non-traditional builders who want to go from idea to working app using AI coding agents.
OpenAn honest comparison of Codex, Claude Code, SSH/tmux, and remote relay workflows for builders choosing a phone-first AI coding setup.
OpenThis series is for PMs, operators, and startup-minded builders who want to turn ideas into working software. Each piece explains who it is for, what problem it solves, and what concrete shipping skill you leave with.
Pick one real use case, choose one AI coding tool, and ship a first usable result in 48 hours.
Focus: tool choice · first ship · confidence
Use AI for first drafts, then tighten the spec with constraints, success metrics, edge cases, and experiments.
Focus: intent clarity · constraints · outcomes
Move from demo-quality output to real code, internal dogfooding, usage data, and launch decisions.
Focus: dogfood · feedback loops · launch criteria
Use your phone as the control plane and your Mac as the execution host for coding, testing, and deployment.
Focus: async control · remote deploy · momentum
As AI makes building easier, the edge shifts to choosing what to build, explaining it clearly, and getting distribution.
Focus: curation · leadership comms · network
Architecture, SwiftUI client flow, JSON-RPC bridge, Tailscale networking, OTA installs, and the phone-first iOS development workflow.
OpenA practical path for PMs, operators, and non-engineers: learn specs, prompts, iteration loops, and how to ship without tooling anxiety.
OpenTool-by-tool comparison with strengths, limits, and the workflows best suited to Codex, Claude Code, Copilot-style tools, SSH/tmux, and relay setups.
OpenThese are weekly build notes, workflow improvements, and lessons learned from shipping iOS and web products with AI coding tools. It is the operating log behind the site.
I’m Chetan Ankola, a product manager and former iOS engineer sharing how I use AI coding tools to build and ship native iOS apps.
My current focus is Codex Relay: a native SwiftUI iPhone app that controls Codex on my Mac so I can build, review, and deploy apps directly from my phone.
This site documents the workflows, architecture, PM frameworks, and real product experiments behind that process. If you are a PM, builder, indie hacker, or iOS developer learning AI-assisted development, this site is for you.
Join builders and PMs learning how to ship real apps faster with phone-first workflows, architecture breakdowns, and practical AI coding lessons.