Validated learning in Agile Hardware Development

Eelco Rustenburg and Gladwell Academy hosted a 2-hour meetup on ’Scaling Agile for Hardware’. With 2.5 years as a lean-agile coach in a hardware-focused setting, I no longer see any reasons why a lean-agile approach should not be extremely beneficial for hardware product development. It was reassuring to find Eelco sharing the same insight after over a decade of coaching in hardware-intensive transformations.

When discussing the second value of the agile manifesto, I typically replace “software” with “solution”. Eelco Rustenburg offers an insightful twist: “Validated learning over comprehensive documentation”, aligning closely with set-based design principles and how we approach feature slicing. Often, our focus isn’t solely on launching product features but on validating learning through Exploration Features, aiming to learn efficiently through minimal viable experiments – or, as Eelco phrases it, “The stupidest thing we can do to learn the most?”. The meetup also provided several other valuable insights through memorable one-liners.

• Move your decision points from before to after, it is an empirical system.
• Slicing of features is a contact sport, bringing people with different perspectives into the room.
• Keep your decision points (not gates), and go through all of them every PI.

SAFe Now

Three major framework updates were unveiled at the conference.

The Operating Value Streams for Large Solution Development article explores two patterns for organizing value streams in large-scale solution development, such as trucks and spacecrafts, using SAFe. The “Nested Value Stream” pattern groups Large Solution Trains sharing the same strategy, funding, and end-customer solution, optimizing for speed. In contrast, the “Value Stream Network” pattern comprises more independent trains with their own intents, roadmaps, and lifecycles, often in a provider-consumer relationship, optimizing for economies of scale. These patterns draw on real-world customer experiences and the broader lean-agile knowledge base.

The updated Artificial Intelligence in SAFe article outlines strategies for becoming an AI-centric company to enhance customer value and workforce augmentation with AI. It details how AI tools can benefit every SAFe role, emphasizing four key areas for accelerating business outcomes: intelligent customer solutions, optimized value stream flow, enhanced customer insights, and increased workforce productivity. Highlighting the importance of AI, it asserts, “AI won’t replace you. A person using AI will,” underscoring that this principle applies to companies and organizations alike.

Measurable Business Outcomes with SAFeoutlines a new workshop toolkit by Scaled Agile designed to align lean-agile transformations with concrete business results. It facilitates collaboration between main stakeholders and the Lean-Agile Center of Excellence (LACe) to ensure strategic alignment up to the board level. Currently, in beta, We Are Movement is authorized to facilitate this promising workshop.

SAFe Next

Chief methodologist Andrew Sales outlined the vision for the next version of SAFe, emphasizing simplification to appeal to the late majority. He highlighted the need to make the framework more accessible by breaking down complex problems into smaller, manageable parts. The aim is to streamline the Big Picture/Overview, with a comprehensive body of knowledge organized in articles linked by a coherent narrative. This approach ensures detailed knowledge is available without overwhelming users.