GenAI PMagentic engineering
A workflow for using AI agents to plan, build, test, and update software with minimal manual intervention. The newsletter treats it as a practical product-development paradigm.
Key Highlights
- Agentic engineering is a software development approach where AI agents help plan, build, test, and update products with minimal manual intervention.
- The concept gained clearer definition through Simon Willison’s writing and discussion of both best practices and anti-patterns.
- For AI PMs, agentic engineering is especially relevant for faster prototyping, better experimentation velocity, and leaner execution.
- Structured workflows such as Compound Engineering’s plan/work loop show how agents can handle complex implementation tasks with greater autonomy.
- The main PM challenge is balancing speed with safeguards like human review, test coverage, and deployment controls.
Overview
Agentic engineering is the practice of building software with AI coding agents that can not only generate code, but also execute tasks, inspect outputs, iterate on plans, and help move work from idea to implementation with limited manual intervention. In the newsletter, it appears as a practical product-development paradigm: a way to use agents to plan, build, test, debug, and update software systems faster than traditional manual workflows.For AI Product Managers, agentic engineering matters because it changes the shape of product execution. Instead of treating AI purely as a chat interface or feature layer, teams can use agents as active contributors in the development loop. That creates opportunities to accelerate prototyping, reduce the cost of experimentation, and expand what smaller teams can ship—but it also introduces new needs around review, reliability, safety, and workflow design.
Key Developments
- 2026-03-05: Simon Willison highlighted anti-patterns in agentic engineering, emphasizing behaviors to avoid, especially letting unreviewed agent-generated code be pushed to collaborators.
- 2026-03-15: Simon Willison discussed agentic engineering in a fireside chat at the Pragmatic Summit in San Francisco, hosted by Eric Lui from Statsig, signaling growing interest in the concept among practitioners and product leaders.
- 2026-03-16: Simon Willison published a definition of agentic engineering as software development assisted by coding agents that can both write and execute code, helping formalize the term and its scope.
- 2026-06-15: Matt Van Horn demonstrated a highly practical workflow using Compound Engineering’s `/c plan` and `/c work` loop to have agents generate step-by-step plans, reverse-engineer web APIs via HAR files, and build SQLite-backed CLIs and agent skills such as Printing Press with minimal direct code reading.
Relevance to AI PMs
- Reduces prototype-to-product cycle time: AI PMs can use agentic workflows to turn ideas into working internal tools, demos, and testable product concepts much faster, especially when paired with coding agents like Claude Code or Codex.
- Changes how teams scope and supervise work: PMs need to define tasks in ways agents can execute well—clear goals, constraints, review checkpoints, and acceptance criteria become even more important than in traditional software workflows.
- Introduces new operational risks and controls: Agentic engineering can accelerate output, but PMs must design guardrails for code review, test coverage, permissions, and deployment policies to avoid anti-patterns like unreviewed changes or brittle autogenerated systems.
Related
- Simon Willison: A key voice in defining agentic engineering and documenting both its promise and its anti-patterns.
- Coding agents: The foundational tools behind agentic engineering; these agents write and execute code as part of development workflows.
- Pragmatic Summit: A venue where agentic engineering was discussed publicly, showing its relevance to builders and product leaders.
- Statsig: Connected through the Pragmatic Summit fireside chat hosted by Eric Lui, reflecting product and experimentation interest in the space.
- Anti-patterns: Closely related because successful agentic engineering depends on avoiding unsafe or low-discipline workflows.
- Compound Engineering: Provided a concrete workflow example with `/c plan` and `/c work`, showing how structured agent loops can improve execution quality.
- Printing Press: An example of what agentic engineering can produce—agent-generated tools and skills built from APIs, HAR data, and wrappers.
- Claude Code and Codex: Representative coding-agent tools that enable agentic engineering workflows.
Newsletter Mentions (4)
“Matt Van Horn uses Compound Engineering’s slash C plan/C work loop to autonomously generate step-by-step AI agent plans that reverse-engineer secret web APIs via HAR sniffing and build SQLite-backed CLIs and agent skills like Printing Press—all without manually reading code.”
#5 ▶️ How This Non-Technical Founder Mastered Agentic Engineering in 50 Minutes | Matt Van Horn Peter Yang Matt Van Horn uses Compound Engineering’s slash C plan/C work loop to autonomously generate step-by-step AI agent plans that reverse-engineer secret web APIs via HAR sniffing and build SQLite-backed CLIs and agent skills like Printing Press—all without manually reading code. Compound Engineering’s `/c plan ` command writes a detailed execution plan to prevent agent laziness and `/c work` executes the plan; Matt used this loop to build and update his Agent Cookie tool in minutes without viewing the plan file. Printing Press ingests official CLIs/APIs, HAR-sniffs secret web APIs (e.g., Kayak Direct, Google Flights), incorporates GitHub community wrappers (e.g., Python Domino’s pizza API), and creates an SQLite database with power-user personas to generate a CLI plus Hermes, OpenClaw, Claude Code, and Codex agent skills—e.g., `pp flight goat` returns cheapest long-haul flights like London at $1,200 per passenger.
“What is agentic engineering? - Defines 'agentic engineering' as the practice of developing software with the assistance of coding agents—systems that can both write and execute code—and gives examples and context.”
#5 📝 Simon Willison What is agentic engineering? - Defines 'agentic engineering' as the practice of developing software with the assistance of coding agents—systems that can both write and execute code—and gives examples and context. The piece situates the term within LLM-driven tooling and points to further reading.
“#11 📝 Simon Willison My fireside chat about agentic engineering at the Pragmatic Summit - Simon participated in a fireside chat at the Pragmatic Summit in San Francisco discussing Agentic Engineering, hosted by Eric Lui from Statsig.”
Today's top 12 insights for PM Builders, ranked by relevance from X, LinkedIn, and Blogs. Ramp Ships 500+ Features Using Claude Code #11 📝 Simon Willison My fireside chat about agentic engineering at the Pragmatic Summit - Simon participated in a fireside chat at the Pragmatic Summit in San Francisco discussing Agentic Engineering, hosted by Eric Lui from Statsig.
“Simon Willison Anti-patterns: things to avoid - A guide section describing behaviors that are anti-patterns in agentic engineering, with emphasis on avoiding unreviewed code being pushed to collaborators.”
#11 📝 Simon Willison Anti-patterns: things to avoid - A guide section describing behaviors that are anti-patterns in agentic engineering, with emphasis on avoiding unreviewed code being pushed to collaborators.
Related
An AI coding assistant environment used for running evaluation skills and agentic workflows. In this issue it is mentioned as a runtime for ai-evals-course material and as an agent in an OpenRouter-like system.
An AI coding agent or environment mentioned as a place to run AI eval skills. It is also listed as one of the agents that can be compared in a shared environment.
A prominent AI blogger and commentator referenced in connection with an article on token reselling and fraud. He is cited as the source of the newsletter item discussing the marketplace and API-key abuse.
Autonomous software agents that write, maintain, and redesign code systems. For PMs, they represent a shift in how engineering and research work gets allocated.
A plugin/pattern used to manage build loops and goal-driven agent workflows. Here it is tied to Codex Desktop and the LFG loop for prototype completion.
Stay updated on agentic engineering
Get curated AI PM insights delivered daily — covering this and 1,000+ other sources.
Subscribe Free