Chapter 3: The Learning Engine

In 2016, while working as a Community Manager at Devcenter.co, I noticed something fascinating. Developers who would start by asking simple questions were, within surprisingly short periods, helping others solve complex problems. This organic evolution of knowledge wasn't just impressive — it raised an important question: Could this natural learning process be systematised and accelerated?

The Evolution from Learn by Doing

The journey to answer this question began with a simple observation: there was a clear skill gap in the developer community. There were experienced developers and complete beginners, but no clear path between these two states. This led to the creation of Learn by Doing, a programme that combined three key learning approaches:

• Project-based Learning — Real-world projects that developers could contribute to
• Task-based Learning — Breaking down projects into manageable pieces
• Apprenticeship Model — Direct mentorship from experienced developers

The programme included bite-sized explainers attached to small tasks, designed to improve understanding and increase completion rates. Mentors worked one-on-one with learners in private spaces.

But despite initial enthusiasm, I faced a challenge: after about two weeks, learners would drop off. Even attempts to incentivise their contributions didn't work. This led to a crucial realisation — I needed to understand how successful learning systems addressed this problem.

Integration of Proven Learning Methods

Research into successful learning systems revealed two key examples:

1. Seth Godin's altMBA, with its remarkable 96% completion rate
2. Josh Kaufman's approach to learning anything in 20 hours

These insights led to a fundamental shift in thinking. Instead of focusing solely on content delivery, the system needed to deconstruct skills using first principles, help learners develop self-correction abilities, remove barriers to practice, and create structured practice environments.

Reducing Time To Learn Skills (TTLS)

Building on Kaufman's 20-hour framework, I developed a system to reduce the Time To Learn Skills (TTLS). My approach focused on four key elements:

Skill Deconstruction — Using first principles to break down complex skills, creating mental models for each component, and making learning paths clear and accessible.

Self-Correction Tools — Microlearning principles for easier assimilation, clear feedback loops, and structured progression paths.

Barrier Removal — Personalising learning paths, matching content to natural affinities, and creating accessible entry points.

Practice Optimisation — Learning Labs for hands-on experience, sprint-based learning cycles, and real-world project integration.

The results were remarkable. In early pilot workshops from March to May 2018, participants were able to learn and apply new skills in just 2–3 hours — a dramatic reduction from the original 20-hour framework.

Case Study: Training at African Fintech Foundry

The real validation of this system came through its application at African Fintech Foundry (AFF). After organising their first demo day and exhibition event, I was invited to train their first Fintech Accelerator Programme cohort.

The training focused on five key thinking frameworks:

• Design Thinking — For understanding user needs and creating innovative solutions
• Business Thinking — For developing sustainable business models
• User Thinking — For deeply understanding user psychology and behaviour
• Market Thinking — For aligning products with market opportunities
• Project Thinking — For effectively executing and delivering solutions

One particularly powerful insight emerged around market-product fit. Participants realised they needed to reverse their thinking — instead of building a product and finding its market, they needed to understand the market first and let it inform their product development.

The effectiveness of this approach wasn't just in the concepts taught, but in how quickly participants could grasp and apply them. The mental models I'd developed allowed complex ideas to be understood and implemented rapidly, validating the TTLS reduction approach in a real business context.

Case Study: Maliyo Games' Learning Approach

The success at AFF caught the attention of Hugo Obi, founder of Maliyo Games. As his team prepared to launch a new fintech product, he saw the value in the learning system. Despite initial reluctance and high pricing on my part, Hugo insisted on bringing this training to his team.

The Maliyo Games experience proved particularly valuable because the team provided direct feedback on improving the training, their suggestions helped refine and enhance the mental models, their success demonstrated the system's applicability across different domains, and Hugo's experience as an entrepreneur provided valuable insights into how to productise and scale the training system.

Recognising Natural Learning Patterns

As I continued to develop and refine the TTLS approach, I began to notice something fascinating: this process wasn't entirely novel. In fact, it was already occurring naturally in many learning environments, just without the systematic framework.

Two personal experiences vividly illustrated this insight. The first came during my journey of learning to skate. As a complete novice terrified of falling, I was introduced to a guide who embodied the essence of smart micro-content and mental model creation. His instructions were precise and actionable: "Bend your knees slightly when standing to maintain balance," "Move your hips to create momentum," "If you feel like you're falling, lower your centre of gravity."

Each of these instructions was a miniature mental model — a bite-sized piece of knowledge that transformed a complex skill into manageable, actionable steps. Within a short time, I was moving on skates, not expertly, but significantly better than when I started.

A similar experience unfolded when I learned salsa dancing. The instructor broke down the intricate dance into simple, digestible steps: "1, 2, 3... 5, 6, 7" — basic forward and backward movements, side-to-side transitions. Each step was a piece of micro-content, and as these pieces were combined, they formed a comprehensive mental model of salsa dancing.

What struck me was how these learning experiences mirrored the systematic approach I had developed, but occurred entirely organically. The key difference was awareness and intentionality. Most people were already using these learning techniques instinctively, but without a structured framework to understand and optimise them.

The crucial insight: effective learning is already happening all around us. Our task is not to reinvent learning, but to understand, systematise, and optimise the natural learning processes that humans have always used.

The Future of Learning Systems

This evolution reveals several key principles for building effective learning systems:

Focus on Mental Models — Break down complex skills into understandable components, create clear frameworks for thinking, and enable quick application of concepts.

Optimise for Speed — Reduce time to competency, remove unnecessary barriers, and create clear action paths.

Integrate Real Feedback — Learn from actual implementation, adjust based on user experience, and continuously refine approaches.

Enable Natural Progression — Match learning to natural abilities, create clear advancement paths, and support organic growth.

The communities that thrive will be those that can effectively integrate these principles into their learning systems. The key is not just in having a learning engine, but in ensuring it's optimised for both speed and retention.

In the next chapter, we'll explore how to scale these learning systems while maintaining their effectiveness and core principles.