When I reflect on my journey in AI and machine learning, I notice a significant divide between what I learned in formal education and what actually proved valuable in the industry. While academic programs excel at teaching mathematical foundations and theoretical concepts, I’ve found that some of the most career-accelerating knowledge comes from unexpected sources.
Throughout my time leading machine learning teams at Google and LinkedIn, and now as VP of Engineering & Data Science at HealthRhythms, I’ve recommended certain books to engineers I’ve mentored—books that have consistently helped bridge the gap between academic knowledge and practical success.
What I Found Missing in My AI Education
My university training provided an excellent foundation in algorithms and techniques, but as I moved into building real-world AI systems, I discovered knowledge gaps that weren’t addressed in my coursework. After guiding multiple AI products from conception to international launch and mentoring engineers making this same transition, I’ve identified several books that provided what my formal education didn’t.
These aren’t typical AI textbooks with complex mathematical notations or the latest neural network architectures. Instead, they address areas I found crucial in my own career: human collaboration, practical implementation, and system design.
Three Books that Transformed my Career in AI
1. How to Win Friends and Influence People by Dale Carnegie
Why it helped me: Despite being published decades ago, this book transformed how I approach teamwork in technical environments. Every significant AI project I’ve led required collaboration across diverse teams. The ability to effectively communicate complex ideas, negotiate for resources, and build consensus around technical decisions often determined project success more than technical brilliance alone.
I’ve found Carnegie’s principles particularly valuable when:
- Presenting technical concepts to non-technical stakeholders
- Building support for new AI initiatives
- Managing cross-functional teams with competing priorities
As the saying goes: “If you want to go fast, go alone. If you want to go far, go together.” In my experience, AI work requires both speed and distance—making these interpersonal skills invaluable.
2. AI Engineering by Chip Huyen
Why it helped me: The gap between a working notebook model and a production AI system was much larger than I initially anticipated. While my early focus was often on improving model accuracy by small percentages, in industry I faced entirely different challenges:
- Building features that could be computed in real-time
- Designing monitoring systems to detect model drift
- Handling training/serving skew effectively
Huyen’s book connected theory to practice for me, covering engineering practices that transform interesting models into valuable products. After seeing many promising AI initiatives struggle not from poor algorithms but from implementation challenges, this knowledge proved essential for my teams.
3. Designing Data-Intensive Applications by Martin Kleppmann
Why it helped me: I discovered that AI systems are fundamentally data systems. Throughout my career building and leading data science teams, understanding the infrastructure underlying our models became just as important as the models themselves.
Kleppmann’s book gave me crucial insights about:
- Database systems storing our training data
- Distributed computing approaches for large-scale model training
- Data processing pipelines feeding our models
- Consistency and reliability considerations for production AI
When my team redesigned our data warehouse at Mindstrong, principles from this book helped us achieve 60% cost savings while improving both robustness and data freshness—all critical factors for the ML applications we built.
Learning Through Building
While these books provided an excellent foundation for my career, nothing replaced hands-on experience. I’ve noticed that the most successful junior engineers I’ve hired and mentored all share one trait: they’ve built real projects that tackle meaningful problems.
Approaches I’ve found valuable:
- Building end-to-end ML systems that solve personal problems
- Contributing to open-source AI projects (even documentation improvements)
- Participating in AI competitions with real-world applications
When encountering knowledge gaps, I’ve found these tools helpful:
- Using AI assistants like ChatGPT and Claude to understand concepts
- Developing strong search skills to find relevant documentation
- Joining communities where practitioners discuss implementation challenges
A Personal Formula for Growth
From my experience building and leading AI teams, I’ve developed a mental model for career growth:
Career Growth = Technical Skills × Implementation Knowledge × Communication Ability
I think of this as multiplication rather than addition—weakness in any area can limit overall potential. The three books I’ve shared have helped me and my mentees develop each component of this formula.
My Suggestion for Students
If you’re studying AI and looking to supplement your academic knowledge:
- Consider starting with “How to Win Friends and Influence People” to develop collaboration skills
- Explore “AI Engineering” to understand implementation realities
- Look into “Designing Data-Intensive Applications” to learn about data infrastructure
- Apply what you learn through concrete projects
The technical foundations you’re building in school are valuable, but I’ve found that supplementing them with these broader perspectives can help develop a more comprehensive skill set for industry work.
About the Author: Skylar Payne has led machine learning engineering teams at Google, LinkedIn, and operated at as a technology executive in startups. With experience taking multiple data products from inception to international launch and mentoring numerous ML engineers, Skylar specializes in bridging the gap between theory and practice for AI.