AI-native engineers shift focus from code to context

Dear friends, colleagues and partners, Tomorrow, April 7th @ 6PM I'm sharing everything I learned after going deep into AI-assisted software engineering across ~15 repositories and after months of intensive usage. For a successful AI adoption we need 3 things: 🧠 CAIR — Confidence in AI Results One of the biggest blocker to AI adoption is psychology. I'll show you how to calibrate and improve trust in AI output and why blind confidence kills codebases. 🧩 Context Engineering This is the real skill of working with gen AI. Context is finite memory and how you fill it determines everything. Get it wrong and you end up with context rot, poisoning, distraction, or clash (yes, each of them is a different one 😅). Get it right and you unlock a different level of productivity through its dynamic nature. 🛠️ Tooling Knowing your tools' capabilities and limits is essential for making the right decisions. Each model serves different types of tasks and complexity levels, and choosing wisely while setting everything up correctly can make a big difference. In my presentation I will talk about all three. I've seen AI adoption transform teams and I've seen it sink products due to lack of (self)control and poor tool choices. Let's discuss about both on Tuesday. Come to argue, come to discuss, or just come curious! 🧐 I am happy to see you all very soon! 😊 My friend, Sergiu Limboi, will moderate the discussion. Thanks, Sergiu! 👉 Register here for free: https://shorturl.at/ClFaO

👀 What does it really take to make AI work in software development? My colleague Otto Fischer knows 𝙖 𝙩𝙝𝙞𝙣𝙜 𝙤𝙧 𝙩𝙬𝙤 on how to get the best out of both. 🤌 Join us at CodeBrew 02 tomorrow evening for the finest AI talks, then stay for our signature Good Vibes after-work. ▶️ https://lnkd.in/dYNUiEJx

We're starting a short series on four pain points we keep seeing in engineering teams that rely on AI to write code, and how we're solving them with Predictable Code. The first one is about what happens when ten or fifty developers each run their own AI sessions with their own context. The fragmentation is invisible until it isn't, and by then a lot of hours have already disappeared into inconsistency. https://lnkd.in/e8yp4gfu #FormalVerification #DeveloperTools #AICodeGeneration

One of the reasons we started Predictable Machines, Inc. was a pattern I kept seeing in teams adopting AI tools: every developer is more productive in their own session, but as a team, the codebase quietly drifts. I wrote the first piece of a short series on the four pain points we're solving with Predictable Code. This one is about the team effect.

Meghan Balavender, Delivery Leader, Technology at The Planet Group, is featured in Spiceworks. She shares great perspectives on how generative AI roles are evolving—and what it really takes to turn AI into business impact. Worth the read.

𝗔𝗴𝗲𝗻𝘁𝗳𝗼𝗿𝗰𝗲 𝗪𝗼𝗿𝗹𝗱 𝗧𝗼𝘂𝗿 𝗖𝗼𝗽𝗲𝗻𝗵𝗮𝗴𝗲𝗻, 𝗦𝗲𝘀𝘀𝗶𝗼𝗻 𝟱: 𝗩𝗶𝗯𝗲-𝗰𝗼𝗱𝗶𝗻𝗴 𝗺𝗶𝘀𝘀𝗶𝗻𝗴 𝗹𝗶𝗻𝗸: 𝘁𝗵𝗲 𝘀𝗲𝗰𝗼𝗻𝗱 𝗯𝗿𝗮𝗶𝗻 Everyone talks about AI coding. Few talk about how to make it scalable, reliable, and useful inside real organisations. That is the missing link. In Session 5, Enrico Maria Dal Compare shares a consultant’s perspective on the hype around vibe-coding, how it can deliver consistent quality, and what a scalable future could look like inside the Salesforce ecosystem. 23 April · 15:30–16:15 Y4 Theater, Copenhagen Join the final session of the day. #Y4 #AgentforceWorldTour #Salesforce

If humans have a future coding along side AI, there’s a whole discipline I think that needs to emerge. Infrastructure for how humans think alongside agents. It’s so exciting for people (me) to be able to produce at this velocity and I think we’re hoping that the models just get good enough that it works at scale because right now we have people vibe coding small things, but the coherence falls apart at scale. Maybe something like coherence management at AI-assisted velocity

As AI generates code at an unprecedented pace, what do developers bring to the table? Watch our latest video to discover how system design essentials remain crucial in today's fast-paced development landscape: https://lnkd.in/gXMx7sQc

I tried using Claude for real coding workflows. Not just “write a function”… But actual development tasks. Here’s what I observed: Claude is strong at: – understanding large codebases – writing structured, readable code – explaining logic clearly But breaks when: – context becomes fragmented – requirements are ambiguous – multiple steps aren’t orchestrated properly That’s when I realized: The problem isn’t the model. It’s how we design the system around it. Good AI usage = prompts Great AI systems = orchestration + context + control Still experimenting with where Claude fits best in dev workflows. #buildinpublic #claude #ai #softwareengineering

Maggie Appleton does a great job covering an emerging challenge of AI development. Because AI has made coding faster/cheaper, "can we build it?" is being replaced by "should we build it?" and "do we actually understand what the customer needs?" The video is technically a GitHub ACE product demo, but don't let that stop you. Watch the first 5 minutes at minimum. https://lnkd.in/gfTZgR3e