matthewsinclair.com · Intelligence. Innovation. Leadership. Influence.

Source: Photo by Jon Tyson on Unsplash

This month’s Engineering Leadership Reading List kicks off by looking at AI’s role in engineering management. Phil Calçado’s deep dive examines back-end architecture for AI-driven products. The piece recounts the development of an AI-powered engineering assistant and the subsequent creation of Outropy, highlighting the technical and operational challenges of scaling AI systems while balancing user needs.

Despite the growth of AI, its limitations in team building are emphasised in Stack Overflow’s analysis. The article argues that while AI can generate code, it cannot replace the foundational learning and system understanding gained through junior engineering roles. Hiring junior developers remains critical for long-term sustainability and talent development in software organisations.

A shift away from traditional OKR frameworks is proposed in Jessitron’s critique. The article challenges rigid cascading OKRs, advocating instead for a model where teams articulate how their objectives contribute to broader strategic goals rather than simply aligning through hierarchical metrics.

A related organisational shift is highlighted in a study on Gen-Z professionals, which finds that many younger workers are avoiding middle management roles due to high stress and perceived lack of value. This reflects broader trends in companies like Meta that are streamlining management layers, prompting leaders to reconsider how management roles are structured.

Structural changes in engineering management also appear in an analysis of corporate downsizing. The piece explores how companies are reducing management layers to cut costs and increase agility, while balancing the risks of organisational inefficiency.

The concept of technical debt is examined from two angles. Aviv Ben-Yosef argues that the term is often misapplied, urging leaders to prioritise only necessary maintenance rather than chasing perfection. Meanwhile, an alternative perspective suggests that bad code is more like an unhedged call option than debt, carrying unpredictable risks that may result in costly rewrites if ignored.

For hiring managers, LeadDev’s guide outlines red flags to identify during interviews, both from the candidate and company perspective. A well-structured interview process benefits both parties, helping ensure alignment between expectations and company culture.

Steven Sinofsky’s reflection on failed engineering strategies explores why many appealing-sounding technical ideas—such as excessive modularity or forced API exposure—often backfire in practice. The piece serves as a cautionary guide to evaluating well-intentioned but impractical engineering decisions.

Lastly, individual career growth remains a key theme, with a discussion of reputation-building in large tech companies. The article explores how an engineer’s early successes shape long-term opportunities and why reputational momentum can be difficult to reverse.

As always, the Quantum Fax Machine Propellor Hat Key will guide your browsing. Enjoy!

right, fit

4-out-of-5-hats Building AI Products—Part I: Back-end Architecture: The article discusses the journey of developing a successful AI-powered assistant for engineering leaders that attracted over 10,000 users within a year, outperforming established competitors. As the AI assistant gained popularity, the article explores the demand for the underlying technology, leading to the creation of Outropy, a platform for building AI products. Key challenges and innovations in structuring AI applications, deploying agents, memory management, and scaling are shared, alongside insights into the balance between technical complexity and user needs.

#AI #SoftwareEngineering #Innovation #Startup #TechTrends

left, fit

3-out-of-5-hats Generative AI is not going to build your engineering team for you: The blog post titled “Generative AI is not going to build your engineering team for you” delves into the intricacies of software engineering and its evolving landscape. It emphasizes that while generative AI can quickly generate a large amount of code, its utility falls short in replacing the foundational work of junior engineers, particularly in understanding, maintaining, and managing complex systems. The article argues that coding is the easiest part of software development, with the real challenges lying in code management and long-term system maintenance. It stresses the importance of hiring junior engineers to ensure a future pool of senior engineers, highlighting that AI can’t substitute the human experience and learning essential in software development.

#GenerativeAI #SoftwareEngineering #Coding #TechCareers #FutureOfWork

right, fit

2-out-of-5-hats Cascading OKRs: We can do Better: The article discusses the limitations of cascading OKRs (Objectives & Key Results) within organisations, suggesting that while alignment through OKRs is appealing on paper, it can often lead to inefficiencies and a lack of true interdepartmental collaboration. It argues that organisations should focus on how different departments enable each other rather than just adding up results, highlighting the need for OKRs to reflect a company’s unique strategy. Rather than rigid goal setting, the article suggests OKRs should support strategic alignment, encouraging departments to tell a story of how their objectives contribute to achieving broader goals.

#OKRs #Leadership #BusinessStrategy #OrganizationalDevelopment #SystemsThinking

left, fit

3-out-of-5-hats Building a GenAI + RAG Micro-SaaS Efficiently: Dan Cohen discusses his experience in building a GenAI + RAG micro-SaaS in just 47 minutes, highlighting the rapid pace of AI development. He emphasizes the efficiency of the tool for sales and marketing teams to assess company match to their Ideal Customer Profile, with integrations into Salesforce/Hubspot for streamlined operations. Cohen notes the importance of tools like Lovable and Supabase for front and backend development, and mentions a personal anecdote about debugging due to insufficient OpenAI API credits.

#GenAI #SaaS #AIdevelopment #Sales #Marketing

right, fit

3-out-of-5-hats Red Flags to Watch Out for During Interviews: The article discusses the importance of recognizing red flags during job interviews from both candidates’ and hiring managers’ perspectives. It emphasizes that interviews are a two-way street where both parties need to evaluate each other’s compatibility. Candidates should look for warning signs like a lack of company alignment or a disorganized interview process, while hiring managers should be wary of candidates who show a lack of curiosity, self-awareness, or adaptability. Understanding these red flags can lead to better hiring decisions and job choices.

#JobInterviews #RedFlags #CareerAdvice #Hiring #InterviewPreparation

left, fit

2-out-of-5-hats Conscious Unbossing: A Shift from Middle Management among Gen-Z: A recent study reveals that over half of Gen-Z professionals are reluctant to pursue roles in middle management, with many preferring individual progression paths over managerial ones. Factors contributing to this preference include perceived high stress and low rewards in such roles. This generational shift comes as major companies like Meta have already started eliminating middle management positions, hinting at a broader trend. Employers are thus encouraged to innovate management roles to attract younger talent and prevent future talent shortages.

#GenZ #CareerDevelopment #Management #Leadership #FutureOfWork

right, fit

3-out-of-5-hats Tech Debt isn’t Debt: Aviv Ben-Yosef argues that the term “technical debt” is often misapplied, leading teams to allocate excessive time to unnecessary updates. He suggests that not all outdated code requires immediate attention and that leadership should assess the actual risks and benefits before prioritizing such tasks. He also emphasizes the importance of distinguishing between true technical debt and issues that directly impact user experience, advocating for a more strategic approach to maintenance and improvements.

#TechnicalDebt #SoftwareDevelopment #Leadership #CodeMaintenance #TechStrategy

left, fit

4-out-of-5-hats Bad code isn’t technical debt, it’s an unhedged call option: The article challenges the common notion of ‘technical debt’ by suggesting that bad code is more akin to an unhedged call option than a traditional debt. Compared to a call option, bad code carries unpredictable future risks that may result in significant losses if not managed properly. The metaphor is encouraged as a way to understand the immediate gains and potentially disastrous future costs associated with neglecting proper coding practices in software development.

#TechnicalDebt #SoftwareEngineering #Programming #FinanceMetaphor #CodeQuality

right, fit

3-out-of-5-hats 225. Systems Ideas that Sound Good But Almost Never Work—“Let’s just…”: In his article, Steven Sinofsky shares insights into systems ideas and engineering patterns that often sound promising but fail when implemented. He highlights the phrase ‘let’s just’, noting that it frequently leads to overly complex solutions. By delving into various examples such as making components pluggable or adding APIs, Sinofsky reflects on why these ideas, although appealing, rarely work in practice. The write-up emphasizes the importance of context and understanding the inherent challenges in engineering, making it both a technical and social science.

#SystemsEngineering #SoftwareDesign #OperationalChallenges #EngineeringInsights #TechLeadership

left, fit

2-out-of-5-hats Why are corporations cutting managers?: Corporate America is seeing a reduction in middle and senior management roles, with companies like Meta reducing management layers to streamline operations. These cuts aim to reduce costs and organizational bloat, focusing on core business functions instead. The article explores the tension between cutting too much and not enough, and how corporations are navigating these dynamics during times of fast-rising stock markets and restructuring urges.

#CorporateManagement #OrganizationalChange #CostCutting #BusinessStrategy #Leadership

right, fit

3-out-of-5-hats Ratchet effects determine engineer reputation at large companies: The article examines how engineers’ reputations and career advancements in large tech companies are shaped by a ratchet effect. This process is based on the completion of projects over time, where early successes lead to more challenging projects, gradually building trust and status within the organization. However, quick judgments from executives and the difficulty in reversing a negative initial impression can hinder an engineer’s progress.

#EngineeringCareer #TechCulture #Reputation #Promotion #CareerGrowth

left, fit

3-out-of-5-hats Solo Coaching: Aviv Ben-Yosef emphasizes the importance of self-directed growth for senior leaders, especially when organizational support is lacking. He advocates for setting personal development goals, creating actionable plans with measurable milestones, and actively seeking feedback from peers and subordinates to facilitate continuous improvement.

#Leadership #SelfImprovement #Coaching #ProfessionalDevelopment #TechLeadership

right, fit

5-out-of-5-hats Scientists map the mathematics behind how we create and innovate: Researchers have developed the Edge-Reinforced Random Walk with Triggering (ERRWT) model to study how combining existing elements leads to innovation. By analyzing data from music, literature, and scientific articles, they found that new combinations follow predictable patterns described by Heaps’ law, highlighting the importance of both reinforcement and triggering in creative processes.

#Creativity #Innovation #Mathematics #ERRWT #HeapsLaw

left, fit

3-out-of-5-hats Top Mistakes Made by IT Architects: Bertrand Florat discusses the most common mistakes made by IT architects, drawing from his own experiences and reflections. He highlights behaviours like the ‘Seagull Architect’ who is disconnected from practical needs, and the ‘Factory Worker Architect’ who mechanically churns out designs without real engagement. Florat also warns against the ‘Golden Hammer Syndrome,’ where an over-reliance on a single tool can lead to poor system designs, and emphasises the importance of aligning architecture with business needs rather than personal aspirations.

Other errors include succumbing to vendor influences, neglecting the broader aspects such as legal and functional requirements, and failing to consider system evolutivity or reversibility. Florat stresses the significance of engaging with stakeholders and maintaining documentation to ensure systems are adaptable and aligned with organisational goals. He concludes by encouraging architects to take ownership of their decisions, remain open to feedback, and be willing to address breakdowns in their systems, reinforcing the importance of practical, resilient architecture.

#ITArchitecture #SoftwareDesign #TechLeadership #ITConsulting #ArchitectureStrategy

Regards, M@

[ED: If you’d like to sign up for this content as an email, click here to join the mailing list.]

Originally published on quantumfaxmachine.com and cross-posted on Medium.

QFM051: Engineering Leadership Reading List January 2025

Stay up to date