Delivery Risks; AI & Tech Debt; Encouraging Conflict; AI for Product Leaders
Issue #39 Bytes
🌱 Dive into Learning-Rich Sundays with groCTO ⤵️
Mitigating Delivery Risk in Software Engineering
Software projects often stumble due to hidden risks. 86% face challenges and a shocking 31% are scrapped entirely! Don't let your project be a statistic. Learn how to proactively identify and mitigate these risks, from planning to post-mortem analysis.
Want to keep your project on track and within budget? Dive into this full article by Typo for practical strategies and tools
Article of the Week ⭐
“There is an emerging belief that AI will make tech debt less relevant. […] The opposite is true - AI has significantly increased the real cost of carrying tech debt. The key impact to notice is that generative AI dramatically widens the gap in velocity between ‘low-debt’ coding and ‘high-debt’ coding.”
AI Makes Tech Debt More Expensive
Quality Leads to Speed is a mantra adopted by product companies that focus on high engineering productivity and continuous delivery. The Evan Doyle focuses his sights on AI’s impact on tech debt. His company specialises in categorizing tech debt, sharing his insights on how modules created with AI introduce new kinds of issues to the future landscape of tech.
GenAI can’t handle High Complexity
If you've experimented with AI coding assistants like Cursor or Aider, you've likely noticed their performance hinges on the complexity of your code. In what he calls high-debt environments characterized by intricate control flows and long-range dependencies LLMs often produce less useful responses.
Moreover, a convoluted codebase not only challenges the AI's response generation but also complicates the developer's ability to craft clear prompts and makes the context necessary to provide even a low quality response more expensive (and power hungry).
Make Your Tools Work for You
Humans should do the abstract work, the modularisation and componentization while AI’s can act as assistants in ideation and analysis. The idea of genAI usage that’s trending positively is to use the human experts to bring the product requirements, architecture and refactor the code into such a state that AI’s can work with little error on it for the few tasks they excel at.
AI-Friendly Approaches
Evan suggests a focus on “unblocking” the AI, doubling down on already well established modularization and single-responsibility practices akin to DDD and Microservices to minimise the blast radius and context needed to assertain the output provided by these tools to be accurate.
Other highlights 👇
How to Encourage the Right Kind of Conflict on Your Team
Amy Gallo shares her experience in mediating and coaching teams. Teams make better decisions and form more effective bonds when they feel safe to voice disagreements. This reduces friction and provides a healthy environment for ideas to be tested, discarded and grown without reprisal.
Conflict Can Be Constructive
Not every disagreement is harmful. When managed correctly, conflict can spark innovation by challenging ideas rather than personalities. Think of it as a little engine friction that, when tuned right, improves overall performance.
Focus on Ideas, Not Individuals.
Healthy conflict centers on the exchange of ideas. Keeping discussions data-driven and proposal-focused helps prevent personal attacks and builds trust within the team.
Practical Strategies
Establish Clear Guidelines. Let your team know it’s Okay to disagree. Show them examples of this behavior and modelled responses.
Designate a Devil’s Advocate. Help everyone argue for and against ideas in a depersonalised fashion, even if their own.
Create Structured Discussion Opportunities. Stay calm and focus on the bigger picture.
Intervene When Needed. Frame tentions in a positive light that highlights different tradeofs. This helps avoid unvoiced concerns leading to personal gripes.
Build Psychological Safety. Remain calm and guide discussions in a “gentle whisper”. The team will pick up on any outlier behavior that is tolerated during discussions which can undermine psychological safety long-term.
A Week in My Life as a Product Leader with AI
Peter Yang shares his weekly workflow. Throughout the week his role as product leader took advantage of AI’s for the following activities:
Summarise customer feedback
Conduct market research. Perplexity with R1 and OpenAI allows for competitive analysis and tech stack comparisons.
Draft a PRD (Product Requirements Document). Turn notes into structured tickets.
Explore design variations
Review tech spec
Make technical trade-offs
Tweak product copy
Improve strategy doc. “An adjacent team is revamping its strategy and wants my feedback. I think the doc focuses too much on tech instead of user problems. I ask AI to suggest edits to make it more focused on customer needs”
Draft customer interview questions
Brainstorm with AI voice
Improve a prompt for an AI product
Edit evaluation criteria
Run LLM evaluations. Common with RAGs and multi-model actors, you can use specialised models to improve prompts for focused tasks.
Share product update. Shape reports into the desired platform’s format (e.g. email → slack, notes → JIRA)
Improve self review
Edit peer feedback
Synthesise interview notes
Find yourself 🌻
That’s it for Today!
Whether you’re innovating on new projects, staying ahead of tech trends, or taking a strategic pause to recharge, may your day be as impactful and inspiring as your leadership.
See you next week(end), Ciao 👋
Credits 🙏
Curators - Diligently curated by our community members Denis & Kovid
Featured Authors - Evan Doyle, Amy Gallo,
Sponsors - This newsletter is sponsored by Typo AI - Ship reliable software faster.
1) Subscribe — If you aren’t already, consider becoming a groCTO subscriber.
2) Share — Spread the word amongst fellow Engineering Leaders and CTOs! Your referral empowers & builds our groCTO community.