More Goto tech episodes

Tech Truth: Agile Evolution & the Future of SW Engineering  Martin Fowler & Kent Beck thumbnail

Tech Truth: Agile Evolution & the Future of SW Engineering Martin Fowler & Kent Beck

Published 2 Jun 2026

Duration: 00:53:34

Software development practices, structured communication methods, Agile framework critiques, AI integration challenges, evolving programming approaches, and the necessity of adaptation, mentorship, and balanced automation-human oversight.

Episode Description

This conversation was recorded at GOTO Copenhagen 2025. https://gotocph.com Martin Fowler - Pioneer of Various Topics around Object-Oriented Technolog...

Overview

The podcast explores a range of topics in software development, emphasizing innovative practices, communication strategies, and the evolving role of technology in the field. It discusses structured communication methods, such as finger-raising techniques and colored card systems, to manage chaotic meetings and prioritize speaker turns, while noting challenges in scaling these approaches to large groups. Remote collaboration tools like Zoom were adapted during the pandemic to replicate in-person structures, underscoring the need for organized virtual communication. The text also examines the practical use of Google Sheets for real-time collaboration and critiques the reliability of AI tools in programming, such as Genie, which often produce inconsistent or misleading results. It highlights the importance of human oversight to validate AI outputs and the risks of overreliance on automated code generation.

The discussion expands to broader themes, including the critique of Agile methodologies at scale, the limitations of frameworks like SAFe, and the need for context-specific adaptations rather than one-size-fits-all solutions. It reflects on the evolving perspective of developers, from junior focus on novelty to senior considerations of feasibility and cost, alongside the challenges of leadership and technological change. The role of mentorship, the potential impact of AI and large language models on programming, and the value of experimentation with new languages and techniques are also covered. Key takeaways emphasize the centrality of communication between developers and stakeholders, the importance of domain understanding, and the need to balance innovation with rigor as the field continues to evolve.

What If

  • What if you implemented a digital "yellow card" system for virtual team stand-ups using Zoom's reaction features?

    • Move: Use Zoom's "Raise Hand" feature as a digital yellow card to queue speaking turns in remote meetings.
    • Why Now? With remote work, chaotic virtual meetings are common, and structured communication tools are critical for focus.
    • Expected Upside: Improved meeting efficiency by reducing interruptions, ensuring all voices are heard, and fostering respect in team discussions.
  • What if you created a peer-review loop for AI-generated code to counteract unreliability?

    • Move: Require all AI-generated code to pass a "Genie Checklist" (e.g., run unit tests, compare outputs across AI prompts, and manually verify logic).
    • Why Now? AI tools like GitHub Copilot are increasingly used, but their outputs are error-prone and unpredictable.
    • Expected Upside: Reduced risk of introducing subtle bugs, faster debugging cycles, and increased trust in AI as a tool rather than a replacement.
  • What if you designed a custom Agile workflow for your solo projects, tailored to your team size and goals (if any)?

    • Move: Build a lightweight Agile framework using Google Sheets for task tracking, with biweekly sprints and user feedback loops.
    • Why Now? Scaling Agile frameworks like SAFe are misapplied, but small teams can adapt principles (e.g., communication, incremental delivery) to their unique needs.
    • Expected Upside: Streamlined project management, better alignment with user goals, and reduced burnout through smaller, achievable milestones.

Takeaway

  • Implement structured communication protocols in meetings
    Use tools like Zoom's hand-raising feature or a shared document to enforce order, inspired by the finger-raising technique and colored card systems discussed. This ensures clarity and respect during discussions, even in large teams or remote settings.

  • Validate AI-generated code rigorously
    Run the same AI query multiple times with slight variations to detect inconsistencies and verify outputs against requirements. Avoid over-reliance on AI for deterministic tasks (e.g., code counting) without manual validation, as errors can be subtle and hard to detect.

  • Adapt Agile principles to your specific workflow
    Prioritize communication with stakeholders and users over rigid frameworks. Tailor Agile practices to your context, focusing on incremental improvements (e.g., shifting from monthly to biweekly integration cycles) rather than adopting generalized scaling methodologies like SAFe.

  • Use real-time collaboration tools for remote coordination
    Leverage tools like Google Sheets for tracking progress, prioritizing tasks, or managing shared data. This minimizes misalignment in remote workflows and ensures all contributors have access to up-to-date information, even if voting features are not used.

  • Incorporate deliberate learning breaks into your development cycle
    After completing a feature, take time to refactor, explore alternative implementations, or study domain-specific knowledge. This prevents burnout and deepens your understanding of both code and user needs, aligning with the "learning between features" approach.

Recent Episodes of Goto tech

10 Jul 2026 The Architect's Guide to the AI Era Luca Mezzalira & Teena Idnani

"AI accelerates coding and research but requires human expertise for system design, governance, and business alignment, with architects balancing innovation, ethics, and practical constraints to ensure AI augmentsnot replacestraditional engineering."

3 Jul 2026 Model-Based Systems Eng. & Requirements Definition Dennis Hansen & Jorge Orellana

Structured system engineering emphasizes sequential layer development from mission definitions to detailed designs, using documentation (e.g., Technical Requirements Reports, SysML) for traceability, iterative refinement, architecture modeling to manage complexity, and collaboration to align technical and managerial priorities while avoiding parallel component development that causes conflicting requirements.

26 Jun 2026 Sovereign Cloud: Who Really Owns Your Infrastructure? Jake Warner & Charles Humble

The evolution of infrastructure management highlights Kubernetes' limitations, introduces Cycle as a decentralized alternative offering flexible, hybrid deployments and data sovereignty compliance, while addressing cost, vendor lock-in, and geographic data privacy challenges through bare metal use, geographic control planes, and expanded provider support with a free developer tier.

19 Jun 2026 Continuous Delivery in a World of Constant Change Abby Bangser & Dave Farley

Continuous delivery principles, AI's challenges in code generation, and the necessity of incrementalism, rigorous testing, and human validation in software development are emphasized, alongside critiques of AI's lack of precision and the push for structured validation in high-stakes systems.

More Goto tech episodes