Agents Don't Do Standups: Building the Post-Engineer Engineering Org — Mike Spitz, PFF

2026-05-15 · Source: AI Engineer · Field: Technology & Digital — Software Development & Engineering, Artificial Intelligence & Machine Learning, Robotics & Autonomous Systems · Depth: Intermediate, long

Summary

PFF, a sports data company serving NFL/NCAA teams and consumers, implemented an AI-driven engineering workflow case study from January to March, involving two top engineers. The initiative aimed to accelerate agent output rather than individual engineer productivity, addressing a bottleneck in their 20-engineer, fully distributed team. This new approach resulted in a 25x increase in deploys, with the two engineers deploying five times daily compared to the larger team's one deploy every five days. Furthermore, their output, measured by blended tickets and code complexity, saw a 10x improvement. The project delivered features in under two months, halving previous four-month estimates, and significantly boosted customer satisfaction scores from an average of 7-7.5 to 8.6 out of 10, indicating a better alignment with customer interests.

Key takeaway

For Directors of AI/ML or AI Architects seeking to accelerate development cycles and improve product delivery, consider a phased adoption of AI agents in your engineering workflow. Focus on automating deterministic, repetitive tasks and integrating agents into design and review processes. This can significantly reduce time-to-market and enhance customer satisfaction, but ensure guardrails are functional and start with your most knowledgeable engineers in non-critical areas.

Key insights

AI-driven engineering workflows can drastically increase deployment frequency and output while improving customer satisfaction.

Principles

• Optimize for agent speed, not just engineer output.
• Customer satisfaction is the ultimate metric for engineering success.
• Automate repetitive tasks to free engineers for complex work.

Method

The process involves huddles for instant feedback, agent-driven spec interviews, lightweight design documents (LDDs), automated ticket/PR creation, and agentic code reviews for style and variable names, replacing traditional Scrum ceremonies.

In practice

• Start AI adoption with non-critical systems.
• Automate QA and aim for self-healing agents.
• Encode organizational coding patterns into composable AI skills.

Topics

• Post-Engineer Engineering Org
• AI Agents
• Automated Software Development
• Lightweight Design Documents
• Agentic Code Review

Best for: Director of AI/ML, AI Architect, Software Engineer

Related on AIssential

• Engineers are becoming advisers to agent-powered work — Agentic AI demands a complete redesign of engineering operations and roles, transforming engineers into strategic advisers to realize its full productivity potential.
• Day 2 - Agentic Engineering/VIbe Coding an AI App from Zero To Production | 🔴 Live — AI agents can rapidly generate functional prototypes and manage project workflows from high-level requirements.
• What Salesforce Learned from 20,000 Enterprise Agent Deployments — Enterprise AI agent success hinges on post-launch iteration, robust guardrails, and deterministic control for reliability.
• Aishwarya Shankar — AI can streamline entire engineering organizations by automating tasks, improving code quality, and enhancing team collaboration.
• Loop Engineering — Loop engineering shifts from direct agent prompting to designing autonomous systems that manage agent interactions and tasks.
• FOD#145: What 100,000 Subscribers Taught Us About the Future of Turing Post — AI development is shifting towards agentic systems, online learning, and efficient, self-improving models.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by AI Engineer.