Shaun Abram
Technology and Leadership Blog
Subscribe to RSS FeedThoughtWorks Technology Radar v34
The latest edition of the always useful Thoughtworks TechRadar, v34, was published back in April. My notes below are mostly just a copy & paste from the original for the sections I found most interesting, mostly for my own benefit. I recommend reading the original from thoughtworks.com or direct from this pdf.
Highlights:
- AI theme continue to dominate with Claude Code and Cursor unsurprisingly make it, firmly, in to Adopt category.
- Model Context Protocol (MCP) as the default integration layer between AI agents and external systems is actively discourage where simpler alternatives exist.
- OpenClaw makes it on to the caution list. “The model requires substantial security trade-offs. The more access you grant it, the more useful it becomes, and the more it concentrates permissions”. A risk to be wary of.
- Claude Code plugin marketplace is on trial as a way to share commands, prompts and skills using a Git-based distribution model.
Tags: radar, techradar, thoughtworks, thoughtworkstechradar
Blocking distracting websites
There are a couple of news and “social” websites that I spend too much time on. I’ve been using tools like Freedom and BlockSite but recently I decided to block some of them completely, at the DNS level.
Tags: distractions, productivity
I’m Begging You: Never Write With A.I., by Bret Stephens in the NY Times
A while ago, I did a quick blog post on my own personal approach to using AI when writing, i.e. don’t start with AI. Instead, start with a blank piece of paper, and while it can be beneficial to use AI to challenge your ideas, don’t copy and paste from it, lest you lose your voice.
AI is always tempting. It can quickly create comprehensive documents. Sometimes too comprehensive. It can be verbose, repetitive, bland. Content that looks contextually appropriate, with the right words in the right order, but adds little actual value. Fluff, you might say. Or AI slop, perhaps, which wikipedia defines as “digital content made with generative artificial intelligence that is perceived as lacking in effort, quality, or meaning”.
So I was intrigued to see this opinion piece in the NY Times (paywalled but trials available), by Bret Stephens, discussing the same topic.
Some highlights below, but I highly recommend reading the article.
Tags: AI, artificialintelligence, nytimes, writing
What is an AI agent?
TLDR: There iss still no standard definition for what an AI “agent” is, but one definition could be:
An AI agent is a LLM based tool capable of autonomous decision making by utilizing tools.
I discuss some of the common characteristics across the emerging definitions of agents and agentic behavior below.
Tags: agent, agentic, agenticcoding, AI, artificialintelligence
Improve Your AI Assistants with Persistent Context
LLM-based assistants, such as Claude, Gemini, ChatGPT and Microsoft Copilot, are wonderful tools and bring a huge amount of utility straight out of the box.
But what if you want the guidance and responses you receive from your AI, to be more customized? To be aware of your specific details and context?
You can give AI context in your prompt, for example:
Assume the role of an expert tax accountant specializing in small business deductions. Answer all my questions concisely, cite relevant IRS rules where applicable, and flag anything I should verify with a professional.
That is a good start. But what if you want such context to persist across multiple conversations? Tools such as Gemini Gems, Claude Projects, Copilot Agents, and ChatGPT GPTs can all help.
Tags: AI, chatgpt, claude, gemini, mscopilot
Research paper summary: Attention is all you need
Attention is all you need is an era defining paper from eight researchers who all worked for Google at the time. The title is a play on the Beatles song, Love is all you need.
The paper addresses some of the shortcomings of other common approaches in Machine Learning (such as recurrent and convolutional layers), in favor of a new architecture, the Transformer. It is based on an approach called self-attention where the model can associate the current token (or word) to any other token in the document, regardless of it’s position. This makes the model so much better at capturing associations and meaning across documents. Further, the Transformer architecture is able to to this processing in parallel, making it faster that preceding models.
I don’t have a Machine Learning background, so this paper was a tough read for me. My interpretation reflects that and I have relied on (and quoted from) other interpretations to help me, including:
- “The Illustrated Transformer” (github.io) by Jay Alammar, which is the most digestible interpretation I could find, and
- This summary from the University of Notre Dame (although exact authorship unclear)
Still, the original is short (16 pages) and is so influential that it was worth reading directly; it came to define the modern era of AI and of LLMs (Large Language Models).
Tags: AI, architecture, artificialintelligence, google, large-language-models, llm, machine-learning, neural-networks
Article summary: Is AI code sabotaging its own ROI case?
Is AI code sabotaging its own ROI case? is an article on leaddev.com where tech journalist Bill Doerrfeld interviews Charity Majors, CTO of Honeycomb.
Some summary notes below, but the TLDR is that the true value of AI in software development isn’t just writing code faster, but accelerating the entire Software Development Life Cycle (SDLC). Ownership is more important than authorship. Keep focussing on the basics of SLOs, stress-testing, canary deployments, feature flags etc. Leaning on Platform Engineering and AIOps to augment operational work. Real ROI won’t come from generating more code faster – it will come from streamlining the lifecycle around it.
Tags: aiops, charitymajors, leaddev, platformengineering, summary
Podcast review: Behind the Craft – Inside Ramp
I found the Behind the Craft: Inside Ramp (spotify.com) podcast an interesting watch. You can find it on Spotify or directly on the Behind the Craft site. The interview is with Geoff Charles (ramp.com), the CPO of Ramp, a fintech platform for corporate cards, and about how they are using AI to build.
I didn’t agree with some of what Mr Charles said, but it was thought provoking.
Some of the funny and/or controversial takes were:
- “a lot of PMS join product management because they might not be good enough at the engineering task”. Wow. That is a controversial opinion, especially from a CPO.
- Geoff suggested leaders should cancel 1-1s to focus on building their API skills. Personally I agree with leaders doing everything can to get up to speed with AI tools, including hands on experience, but not at the expense of neglecting your direct reports & team.
Other highlights…
Tags: AI, artificialintelligence, claude, podcast
Book summary: Made to Stick, Chip & Dan Heath
Made to Stick: Why Some Ideas Survive and Others Die is a book by brothers Chip & Dan Heath. It tries to explain what makes some ideas memorable or interesting. The book has a somewhat similar style to Malcom Gladwell and explicitly references him:
We adopted the “what sticks” terminology from one of our favorite authors, Malcolm Gladwell’s The Tipping Point, which examined the forces that cause social phenomena to “tip,” or make the leap from small groups to big groups.
I liked this book. The SUCCES framework (a Simple Unexpected Concrete Credentialed Emotional Story) is useful enough, but I think I actually liked these 3 topics from early sections the most:
- The Curse of Knowledge: Once we know something, it is hard to imagine what it was like not to know it. Our knowledge has “cursed” us. And it becomes difficult for us to share our knowledge with others, because we can’t readily re-create our listeners’ state of mind. Since we can’t unlearn what you already know, the suggests using the SUCCES framework to transform your ideas to beat the Curse of Knowledge.
- Don’t bury the lead: The first “lead” sentence of a story contains the most essential elements. Don’t allow writing to become so steeped in the details that you lose the core message — what readers will find important or interesting.
- The inverted pyramid: After the lead, information is presented in decreasing order of importance. Journalists call this the “inverted pyramid” structure — the most important info (the widest part of the pyramid) is at the top. The inverted pyramid is great for readers. No matter what their attention span, the inverted pyramid maximizes the information.
Finding the core and writing the lead both involve forced prioritization. The SUCCES framework can help you structure and communicate the points for best effect.
See below for summary notes on each chapter.
Tags: 2026, bookreviews, books, reading
Book review: Focus, by Daniel Goleman
Focus: The Hidden Driver of Excellence (amazon.com), by Daniel Goleman
Overall I found this book just an OK read. Some parts good and interesting. Some parts read like a Malcolm Gladwell book. But ultimately, I just didn’t have any big takeaways. I do find the subject matter interesting, but something about the style or flow just didn’t work for me. The theme is focus, but the book jumps around a lot in a way that left me wondering what the point really was.
I agree with some of the reviews on Goodreads, such as “a mishmash of ideas, supposedly linking all with the focus of focus. It didn’t work for me.” and “Book did not have enough ‘focus’ to hold my interest.”
Still, I made some notes below on the parts that were more interesting to me.
Tags: 2026, bookreviews, books, focus, reading
2025 Books
I read a dozen or so books in 2025, including:
- The Old Man and the Sea, Ernest Hemingway (re-read)
- A Farewell to Arms, Ernest Hemingway
- Tortilla Flat, John Steinbeck (re-read)
- Inner Excellence, Jim Murphy
- The Bonfire of the Vanities, Tom Wolfe (DNF)
- Dead Lions, Mick Herron
- Narrative of the Life of Frederick Douglas, Frederick Douglas
- The Lessons of History, Will & Ariel Durant
- The Richest Man in Babylon, George S. Clason
- Make Your Bed, William H McRaven
- Deep Work: Rules for Focused Success in a Distracted World, Cal Newport
- Co-Intelligence: Living and Working with AI
- Moonwalking with Einstein, Joshua Foer (re-read)
- The Psychology of Money, Morgan Housel
See reviews below, but the highlights from 2025 as a whole are:
- Most enjoyable read: The Old Man and the Sea, but since it was a re-read, l’ll mention “Dead Lions” by Mick Herron too.
- Most inspiring book: Narrative of the Life of Frederick Douglas, by Frederick Douglas
- Best motivational book: Deep Work
See some notes and reviews below.
Tags: 2025, bookreviews, books, reading
Book review: The Psychology of Money by Morgan Housel
The Psychology of Money, “Timeless lessons on wealth, greed, and happiness”(amazon.com) is a book by Morgan Housel, a writer and a partner at a a venture capital firm.
I found this book to be an easy read, reading it over one ski weekend. It’s not a bad read, although for me I felt it underlined some beliefs I already had, rather than changing my thinking much. Entertaining albeit nothing new or groundbreaking. You can see a decent summary here.
Some chapter notes from me below.
Tags: 2025, bookreviews, books, finance, reading
Talk summary: Leading a Team of Developers Using AI by Venkat Subramaniam
I attended NFJS’s Tech Leader Summit conference in Florida last week. One of the talks was “Leading a Team of Developers Using AI” from Venkat Subramaniam.
AI has changed software development irrevocably. We should lean in and play to AI’s strengths, while utilizing the skills that we as engineers still bring to the table. Machines are fast at finding patterns in large data sets, but humans have the critical thinking and intelligence to see beyond existing data sets.
While typing raw code may soon be over, many skills remain essential:
- Problem solving
- Understanding requirements (which may be incomplete and contradictory)
- Handling trade offs
- Communicating
Maintainability, reliability, performance, security and robustness all remain important aspects of software projects that AI may not address sufficiently.
AI has been trained on vast amounts of code written by humans. But much of it is bad (we’ve all written bad code). And now humans complain that AI generates bad code. That is karma!
AI can’t differentiate good code from bad. But you can, especially in the context of your current environments and needs.
Use AI for
- ideas, not solutions
- prototypes, not production code (AI can allow you to bring new ideas to test the market much faster. Similar to how 3D printing did for physical objects)
- finding bugs in and documenting existing code, not writing new code (AI excels at finding issues in complex code that humans struggle to understand)
Tags: conferencetalks, nfjs, nofluffjuststuff, summary, venkatsubramaniam
My personal approach for writing with AI
My own personal notes on using AI when writing. AI is a great tool, but not one to be overused. Avoid creating AI slop. I put these notes together after reading Co-Intelligence: Living and Working with AI by Ethan Mollick, which is a good read.
Tags: AI, artificialintelligence
Book review: Co-Intelligence by Ethan Mollick
We covered Co-Intelligence: Living and Working with AI (amazon.com) by Ethan Mollick at a book club at work, and I found it to be a good read. Part 1 at least, part 2 perhaps less so.
Read on for a review and some notes.
Tags: AI, artificialintelligence, bookreviews, books