By Andrew Park | 2024-10-23
Recently, Curtis Michelson, a lead organizer of Product Tank—a global community of product professionals—asked how Perplexity’s product culture compares with Product Oriented Software Engineering (POSE). His question was sparked by Perplexity’s innovative team structure and processes, and how they align (or diverge) from the principles of POSE, which I’ve outlined in my forthcoming book Product Oriented Software Engineering, set to be published next year. As the author of POSE, I’ve designed this approach to enhance product development by placing Product Management and Design at the core of the process, reducing workload and burnout risk while promoting high-quality, user-centric outcomes. This article will break down the key similarities and differences between these two approaches.
AI-First Philosophy
Perplexity fully embraces an AI-first approach, using AI for rapid decision-making and faster processes. POSE aligns with this philosophy, advocating for AI tools like large language models (LLMs) to be deeply integrated into daily workflows. POSE also provides insights on transitioning employees to widely adopt AI in their roles, encouraging all team members to use AI continuously throughout the day.
Team Organization: Parallelization vs. Alignment Through Understanding
Perplexity reduces coordination costs through extensive parallelization, enabling fast, independent work in small teams. POSE, while also focused on reducing coordination costs, achieves alignment by bridging the chasm of understanding between product management and technical teams. This is accomplished through a combination of automated daily status reports on developer activity and fostering mutual growth in knowledge—technical teams learn more about the ‘what’ and ‘why’ behind product decisions, while product managers gain deeper technical insights. This approach ensures alignment with the product vision, maintaining strategic clarity even as teams and projects scale.
Small Teams vs. Fewer Teams
Perplexity relies on small, highly autonomous teams, which works well in early stages but can overwhelm product managers as the size of software projects scales, requiring multiple software teams. Agile and Scrum were designed with the context of a single team, and scaling frameworks often add more process, mistakenly believing it is the solution. POSE takes a different approach by cultivating strong software craftsmanship skills, enabling fewer, more versatile teams to architect designs for multi-million lines of code that remain comprehensible to the human brains maintaining them. This focus on craftsmanship ensures codebases are both scalable and maintainable as projects grow in complexity.
Maintaining Flat Structures as Teams Scale
Perplexity adopts a flat structure with minimal management, suitable for rapid decision-making but challenging to maintain as the company grows. POSE supports a flat structure but addresses scalability by using tools like PulsePoint to ensure alignment and visibility without adding traditional management layers. This enables sustainable strategic clarity and cohesive outcomes as the organization expands.
Technical PMs and Engineers with Product Domain Knowledge
While Perplexity emphasizes the importance of technical PMs and engineers with strong product domain knowledge, POSE goes further by focusing on mechanisms to cultivate this talent from within. POSE asserts that the future of software development relies on engineers and PMs who understand both the ‘what’ and ‘why’ of product decisions, rather than solely focusing on the ‘how.’ Recognizing that there isn’t enough external talent to meet demand, POSE emphasizes internal growth, equipping existing teams to develop this crucial knowledge. By fostering this mindset, POSE promotes deeper collaboration and strategic alignment between product management and engineering.
Key Takeaways
POSE provides a framework that places Product Management and Design at the center of the software delivery process, greatly reducing the workload for product managers—a field where burnout is common. While both Perplexity and POSE emphasize rapid iteration, minimal bureaucracy, and strong individual contributors, POSE offers a more structured approach to maintain alignment, strategic depth, and scalability. Perplexity’s decentralized model suits early-stage startups, while POSE presents a pathway for scaling complex, high-consequence software products sustainably. Ultimately, POSE offers a comprehensive framework for long-term growth and strategic product alignment.
If you’re a product manager or engineering leader looking to adapt development practices for better product outcomes, check out my book, Product Oriented Software Engineering (POSE). It offers actionable strategies to align product management with engineering efforts. 📘 Learn more here.