The end of product managers? Why LinkedIn is turning PMs into AI-powered “full stack builders”

Key Concepts

Full Stack Builder Model: A new approach to product development at LinkedIn that empowers individuals, regardless of their role, to take an idea from conception to market.
Human + AI Product Team: A collaborative model where humans and AI agents work together to build products, enabling faster iteration, adaptability, and efficiency.
Time Constant of Change vs. Response: The idea that the pace of technological and skill-based change is accelerating faster than organizations can adapt.
Process Complexity: The traditional product development lifecycle has become overly complex with numerous substeps and reviews, slowing down innovation.
Organizational Complexity: The specialization of roles within traditional structures leads to fragmentation and inefficiency.
Core Builder Traits: Vision, empathy, communication, creativity, and judgment are identified as essential human skills that AI cannot yet replicate.
AI Automation: The goal is to automate all tasks outside of these core builder traits to free up human capacity.
Platform, Tools, and Culture: The three key pillars for implementing the full stack builder model.
Change Management: Crucial for successful adoption, requiring incentives, motivation, and examples, not just providing tools.
Associate Product Builder (APB) Program: A new program replacing the APM program, designed to train individuals in coding, design, and product management.
Growth Mindset: Emphasizing continuous learning and improvement over reaching a static state.

Full Stack Builder Model at LinkedIn

The Necessity for Rethinking Product Development

The transcript highlights a critical juncture in the evolution of work and product development, driven by the accelerating pace of change. Tor Cohen, Chief Product Officer at LinkedIn, explains that by 2030, 70% of the skills required for jobs will have changed. This necessitates a fundamental reimagining of how products are built to remain competitive. The traditional product development lifecycle, once a structured process, has become overly complex due to the proliferation of substeps and reviews, leading to organizational bloat and slow iteration cycles. This complexity, Cohen argues, is a symptom of a broken model that needs to be addressed.

The Full Stack Builder Model: Vision and Goal

LinkedIn is experimenting with a new paradigm called the "full stack builder model." The core objective is to empower individuals, irrespective of their functional role (e.g., engineering, product, design), to take an idea from its inception to market launch. This model envisions a fluid interaction between humans and AI, creating a "human + AI product team" capable of rapid adaptation and efficiency. The emphasis is on enabling builders to develop experiences end-to-end, combining skills traditionally siloed in distinct domains.

Core Builder Traits vs. AI Automation

Cohen identifies five key traits that he believes great builders should focus on, as these are areas where human ingenuity currently surpasses AI:

Vision: Formulating a compelling stance about the future.
Empathy: Possessing a profound understanding of unmet needs.
Communication: The ability to align and rally others around an idea.
Creativity: Generating possibilities beyond the obvious.
Judgment: Making high-quality decisions in complex and ambiguous situations.

All other aspects of the product development process are targeted for automation through AI.

The Three Pillars of Implementation

The successful implementation of the full stack builder model relies on three interconnected components:

Platform: This involves significant investment in re-architecting core platforms to enable AI to reason over them effectively. This includes building composable UI components and ensuring the platform is AI-ready. Cohen emphasizes that off-the-shelf third-party tools often require substantial customization to work with a company's unique stack and design systems.
Tools and Agents: This refers to the development of specialized AI agents designed to automate specific tasks. Cohen provides examples of internally built agents:
- Trust Agent: Helps identify vulnerabilities and potential harm vectors in product specifications and ideas, built by the head of trust.
- Growth Agent: Incorporates LinkedIn's unique growth loops, funnels, and past tests to critique and improve product ideas.
- Research Agent: Trained on member personas and historical research data to provide insights and critique marketing specs.
- Analyst Agent: Enables querying the entire LinkedIn graph without relying solely on SQL or data science teams. These agents are currently in an "MVP++" stage, with plans for broader internal rollout.
Culture: This is identified as the most critical and challenging aspect. It involves building incentives, motivation, and demonstrating successful examples to encourage adoption. Cohen stresses that simply providing tools is insufficient; companies must actively manage the cultural shift.

Examples and Case Studies

Open to Work Feature: The Trust Agent was used to review the "Open to Work" feature, identifying not only initial vulnerabilities but also previously missed issues that had emerged later, demonstrating its effectiveness in proactively addressing trust vectors.
Semantic People and Job Search Teams: These teams utilized parts of the full stack builder tools to help build their features, with PMs creating their own dashboards without waiting for design resources.
User Researcher to Growth PM Transition: A user researcher, leveraging the full stack builder tools, successfully transitioned into a Growth PM role, showcasing the program's potential for career mobility and cross-functional development.
Partnerships and BD Teams: These teams are using the tools to build developer portals and connectors, bypassing traditional reliance on engineering resources.

Step-by-Step Processes and Methodologies

Agent Development: The process involves training agents on specific corpora of data, including internal knowledge bases, research, and support tickets. Cohen highlights the importance of curated "golden examples" rather than simply providing access to all company data, as this improves the AI's understanding of context and importance.
Pod Model: Teams are organized into smaller, cross-functional "pods" that tackle specific missions for a defined period (e.g., a quarter) before reassembling into new pods. This fosters nimbleness and focus.
Associate Product Builder (APB) Program: This program involves rigorous training in coding, design, and PM, preparing individuals to become full-stack builders and contribute to pods.

Key Arguments and Perspectives

AI as an Empowering Force: Technology, particularly AI, should be about empowerment, enabling individuals to achieve more.
Necessity of Adaptation: The rapid pace of change in skills and job roles makes adaptation not just beneficial but essential for survival.
Critique of Traditional Models: The current product development process is characterized by excessive complexity, leading to inefficiency and a disconnect between the pace of change and organizational response.
Top Talent Amplification: AI is currently making amazing people even more amazing, as top performers are more inclined to leverage new tools to enhance their craft.
Culture as the Differentiator: While platforms and tools are prerequisites, successful AI adoption hinges on effective change management and fostering a supportive culture.
Permission to Innovate: Individuals should not wait for formal organizational changes but should proactively explore and utilize new tools and methodologies.

Notable Quotes

"The skills required to do your job by 2030 it will change by 70%. So whether or not you're looking to change your job, your job is changing."
"In order to stay competitive, you actually have to go back to some first principles, go back to the drawing board and reimagine what it means to be building."
"It's really a fluid interaction between human and machine."
"The goal itself is to empower great builders to take their idea and to take it to market regardless of their role in the stack and which team they're on."
"It's not enough to give them the tools. You have to build the incentives programs, the motivation, the examples to how you do it."
"Top talent has this tendency of continuously trying to get better at their craft."
"The key trait that I'm emphasizing for builders is... judgment."
"The time constant of change is far greater than the time constant of response."
"The work itself is not complex but the process we made very complex."
"We call it the full stack builder model."
"It's not that you have to break the model. I think the model is broken. It's just this pace of change is is helping us realize it."
"The culture aspect is by far for me the biggest and most important thing to do."
"If you're looking for a formal reorg or declaration to start building differently, you're waiting too long."
"It's about continuously growing and evolving."
"Do not wait. Really focus on the progress you're making."

Technical Terms and Concepts

Time Constant: A measure of the time required for a system to respond to a change. In this context, it refers to how quickly skills and jobs change versus how quickly individuals and organizations can adapt.
Composable UI Components: Building blocks for user interfaces that can be easily combined and reused.
AI Reasoning: The ability of an AI system to process information and draw logical conclusions.
Context Window: The amount of information an AI model can consider at any given time when processing a prompt.
Hallucination (AI): When an AI generates false or nonsensical information.
MVP++: An enhanced Minimum Viable Product, suggesting a functional but not yet fully polished version of a tool or feature.
Orchestrator (AI): A system that coordinates and manages multiple AI agents to work together.
Product Jam: An internal LinkedIn process for defining and refining product ideas.
SQL (Structured Query Language): A standard language for managing and querying relational databases.
360 Feedback: A performance appraisal method where feedback is gathered from multiple sources, including peers, subordinates, and supervisors.
GA (General Availability): The stage when a product or feature is officially released to all users.
PRs (Pull Requests): A mechanism in software development for proposing changes to a codebase.
FOMO (Fear Of Missing Out): The anxiety that an exciting or interesting event may currently be happening elsewhere, often aroused by posts seen on social media.

Logical Connections Between Sections

The transcript flows logically from identifying a problem (rapid change, complex processes) to proposing a solution (full stack builder model). It then details the components of this solution (platform, tools, culture), provides concrete examples, and discusses the implementation challenges and successes, particularly focusing on change management. The discussion on core builder traits versus AI automation sets the stage for understanding the human element in this new paradigm. The introduction of the APB program and the emphasis on culture and incentives directly address how to drive adoption and overcome resistance.

Data, Research Findings, and Statistics

70% Skill Change by 2030: A significant statistic indicating the dramatic shift in required job skills.
Fastest Growing Jobs Growth: The fastest-growing jobs are increasing by over 70% year-over-year, highlighting the dynamic nature of the job market.
70% of Fastest Growing Jobs Not on List a Year Ago: This statistic underscores the emergence of entirely new job categories.
AI Agents Saving Hours: Teams are reporting saving hours of work per week through the use of AI agents.
Maintenance Agent Handling 50% of Failed Builds: A testament to the efficiency gains in the code-to-launch phase.
IDC Study (Vanta Sponsor): Vanta customers slashed over $500,000 a year and are three times more productive.
Miro Survey: 76% of people believe AI can benefit their role, but over 50% struggle to know when to use it.

Section Headings

The Imperative for Change: The Accelerating Pace of Skill Evolution
Introducing the Full Stack Builder Model at LinkedIn
Core Human Traits vs. AI Automation
The Three Pillars: Platform, Tools, and Culture
Internal AI Agents: Examples and Functionality
Implementation Strategies: Pods, APB Program, and Change Management
Measuring Success and Early Wins
The Critical Role of Culture and Incentives
Lessons Learned and Future Outlook
Lightning Round: Books, Media, Products, and Life Philosophy

Synthesis/Conclusion

The LinkedIn full stack builder model represents a proactive and radical approach to product development in the age of AI. By recognizing the accelerating pace of change and the limitations of traditional, complex processes, LinkedIn is investing in a future where empowered individuals, augmented by AI, can innovate more rapidly and adaptively. The model's success hinges on a holistic strategy encompassing platform re-architecture, specialized AI tools, and, crucially, a robust cultural shift driven by incentives, clear communication, and a focus on continuous growth. While the journey involves significant investment and careful change management, the potential for increased agility, resilience, and individual career growth makes this a compelling blueprint for the future of product development. The emphasis on human judgment, creativity, and vision, combined with AI's automation capabilities, suggests a future where human potential is amplified, not replaced.