The coding-agent story became more tangible this week because it moved into a place everyone understands: the phone. Cursor launched an iOS app for guiding a coding agent on the go, according to TechCrunch, while GitHub published an evaluation of its Copilot agentic harness across models and tasks. Add DeepMind's Gemini 3.5 Flash computer-use release, Hugging Face's vLLM Jobs, and Google Research work on accelerating Gemini Nano on Pixel, and the pattern is clear: software work is turning into agent supervision.
This does not mean developers stop thinking. It means the interface to work changes. Instead of sitting at a desk and typing every change, a developer may assign a task, let an agent inspect the repo, review the proposed diff, approve or reject steps, and check the evidence from a phone between meetings. The center of gravity moves from keystrokes to control loops.
GitHub's harness evaluation matters because agents need to be measured as systems, not vibes. A code agent is not just a language model. It is a bundle of context retrieval, tool use, test execution, editing, planning, runtime cost, and failure recovery. If a mobile app makes approval easy but the harness is weak, the result is faster bad decisions. If the harness is strong, mobile supervision becomes a real productivity pattern.
The indie-builder layer shows the missing primitives. Framein points at local work-state context. GSV imagines a personal AI computer. AgentWatch focuses on runtime budget enforcement. LLMSim helps load-test LLM apps. Dribble brings AI into database work. OpenATP explores automated theorem proving in Lean. WtfisMyRepo attacks the everyday pain of understanding a new repository. None of these community projects should be treated as adoption proof, but together they map the problems serious agent workflows need to solve: context, budget, testing, local control, specialized interfaces, and verification.
This is why coding remains the best early lab for agentic work. Code has files, diffs, tests, version control, logs, and rollback. Those tools make agent work inspectable. If the AI changes a function, a human can review the diff and run tests. That is not perfect, but it is clearer than asking an agent to "handle strategy" or "manage operations" with no evidence trail.
The bigger question is what moves next. Once people get comfortable assigning software tasks and approving evidence from a phone, the same supervision pattern can migrate into data analysis, design, finance, legal review, support, and operations. The future of AI work may be less about chatting with a bot and more about checking a queue of machine-generated work. The winning products will not just generate. They will make review fast, safe, and hard to fake.
The best near-term products will probably feel conservative. They will show small units of work, keep risky operations behind approvals, surface test results, and make it easy to roll back. The flashy version is an agent that claims it built the whole feature while you were at lunch. The useful version is an agent that leaves a trail good enough for a tired reviewer to trust at 5 p.m.