The Engineering Leadership & Wellbeing Newsletter 🌱 #7
As you might’ve noticed, we made the leap from Ghost to Substack! 🚀
We did this because we wanted a wider reach and to continue delivering top-notch content on engineering leadership and wellbeing. We’re also hoping to leverage Network Effects, so please share when you have a moment.
As we settle into our new platform, I just wanted to say thank you for sticking with us.
Topics for the day: Toxic engineering manager, Knowledge management, Will AI replace devs? Continuous Integration, Cheating on technical interviews w/ ChatGPT, Errors and success, Performance management, Breaking the stress cycle
Let’s dive right in ♨️
Spotting Toxicity in Your Engineering Manager
Rummaging through my archives, I stumbled upon an old blog post, which seemed to have left an impact on my readers. While we've all encountered toxic engineering managers, this post offers a definitive set of criteria to diagnose the issue once and for all.
Read more: 7 Telltale Signs You Have a Toxic Engineering Manager ☣️
Your Guide to Effective Knowledge Management📚
Sebastien Dubois shares a few secrets to mastering knowledge management with note-taking! He explores everything from the fundamentals and hierarchical structures to naming conventions, turning digital chaos into a goldmine of never-forgotten wisdom.
Read more: The Art of Note Naming: Keys to Effective Knowledge Management
Generative AI Will Not Replace Your Software Developers 🤖🚫
Don’t worry, AI won’t replace your devs anytime soon, according to a recent study. Researchers from Princeton University tested how well giant language models (LLMs) could handle real-world coding problems. And guess what? Turns out, they're not quite ready to steal your thunder! Their custom-built model, SWE-Llama, could tackle only the most straightforward GitHub issues. Meanwhile, big-name LLMs such as Anthropic’s Claude 2 and OpenAI’s GPT-4 barely cracked 4.8% and 1.7% success rates, respectively.
Read more: Can Language Models Resolve Real-World GitHub Issues?
Everything You Need to Know about Continuous Integration 🔧
Continuous Integration involves merging code changes into a shared codebase daily, with each integration automatically tested to detect errors. This practice minimizes delivery delays, streamlines integration efforts, and supports a dynamic codebase for rapid feature development. Martin Fowler has just unveiled his latest updates on CI and it’s well worth the read.
Read more: Continuous Integration by Martin Fowler
Does ChatGPT Make it Easy to Cheat in Technical Interviews? 🎲
Interviewing.io, a platform for engineer interview practice and recruitment, examined ChatGPT's role in interview cheating. Professional interviewers and senior-level job seekers participated, with questions including LeetCode questions and custom queries. Candidates using ChatGPT to cheat performed marginally better, hinting at its potential to impact interview results and urging companies to rethink their question formats.
Read more: How hard is it to cheat in technical interviews with ChatGPT? We ran an experiment
Bonus: Mastering the Engineering Management Interview: 30 Essential Questions + Sample Answers
Errors as Signs of Success 🚩
I enjoyed reading this post by A Smart Bear. It discusses how mistakes could be positive outcomes of good decision-making and progress. It emphasizes the importance of recognizing and embracing these "errors" as natural by-products of growth and innovation. He gives great examples from re-adding features to a backlog, pivoting strategies shortly after creation, and refactoring infrastructure after rapid growth. The concluding paragraph really resonated with me:
Celebrate our devotion to good decisions and good strategy, even when it has negative consequences. That means you have a strategy, and have the ability to make the tough, wise decisions. - A Smart Bear
Read more: The “errors” that mean you’re doing it right
A Performance Management Conundrum 🎭
Who is most capable of judging your performance: your peers or managers? In this post by lcamtuf, the complexities of performance management in large tech companies are explored, shedding light on dissatisfaction with traditional manager-led evaluations. While peer-based systems, pioneered by Google, aimed for fairness, they often result in overly positive feedback and reliance on managers in decision-making.
On Breaking the Stress Cycle 😬
Clinical psychologist, Jenny Taitz, offers these four techniques to help individuals combat Perseverative Cognition.
⚓ Anchor yourself: Ground yourself in the present moment by focusing on your senses and questioning whether your thoughts are helpful.
💭 Take your thoughts less seriously: Use cognitive defusion to create distance from unhelpful thoughts and treat them less seriously.
🌱 Sit with uncertainty: Practice accepting uncertainty and fostering openness to the present moment rather than engaging in endless speculation.
💖 Self-validate: Legitimize your emotions and move past repetitive thoughts by acknowledging and accepting your feelings.
Read more: How to Stop Dwelling on Your Stress
Engineering Leaders to Follow on Substack:
Did you like this edition of ENGINEERING LEADERSHIP & WELLBEING 🌱?
If you're looking for a specific subject to read about, reach out to me at tomek@howareyou.work, and I'll see what I can put together.
Thank you for the mention, Tomasz and great article!
Thanks for the mention Tomasz 🙂