InApp : Tech Radar

We've spent decades pretending that being a good engineer is about how fast you can type or how well you know your framework's syntax. It isn't. The real work has always been the mental heavy lifting, the process of building a precise, working model of a complex domain inside your head before you ever touch a keyboard. Agentic AI is breaking that assumption. We're entering a period where you can generate functional software without ever actually building that mental model yourself. If the machine handles the translation from logic to code, the traditional link between understanding a problem and implementing its solution is gone. This shift changes everything about how we define technical competence. If the implementation is becoming a commodity, we have to figure out what's left for us to do. The Illusion of the Technical Moat The barrier to implementing complex software is dropping because LLMs are great at the "how." Writing a standard CRUD API o...

You know exactly what a dickover is, even if you didn't have a name for it until now. If you spend any time on the internet, you encounter them every day. They are those specific, infuriatingly pedantic corrections that exist solely to make the person writing the post look wrong. The term actually has much deeper roots than a random Twitter argument. It traces back to the Dickover National Hunt race, a fixture in the racing calendar that carries a certain level of prestige. There is a long history of endurance baked into the name, though we've mostly repurposed it to describe the exhausting social friction of the digital age. It’s a weird linguistic pivot. We took a high-stakes sporting event and turned it into a way to describe someone being an insufferable jerk in a comment section. I've been thinking about how that transition happened and why the term stuck so well. The Origins of the Dickover The Dickover is a specific type of long-distance turf race that ex...

We talk a lot about capital efficiency, but we rarely talk about what happens when capital simply stops moving. If investment begins to stagnate and money stops circulating through the ecosystem, the fundamental mechanics of growth might not just slow down. They might break. I've watched enough cycles to know that liquidity is usually the invisible engine behind every "innovation" we celebrate. When that engine stalls, you don't just get a recession. You get a permanent shift in how much risk anyone is willing to take. It's hard to build something new when the only available capital is sitting in a high-yield savings account or parked in legacy assets. The real danger isn't a market dip. It's the possibility that we've entered a period where the cost of stagnation is higher than the cost of a crash. We need to look at where the money is actually going, because the current movement suggests a much more static future than anyone wants to admit. ...

Ed Mansell spent years curating what is likely the largest personal LEGO Star Wars collection ever assembled. It wasn't just a hobby. It was a $200,000 commitment to specific, hard-to-find sets that took a lifetime to track down. When his father's age made it time to move the collection, the plan was simple. Ed and his son Bryan reached out to Bricks & Minifigs Salem-Keizer to facilitate a massive, organized sale. The shop was ready. They even put up the posts to announce the arrival of the hoard. But the collection didn't disappear because of a warehouse fire or a clumsy mover. It was dismantled by a sophisticated retail scam involving fraudulent returns. It’s a weirdly specific way to lose something so tangible. You expect a thief to take the finished models, but instead, a series of manipulated transactions just eroded the collection piece by piece. Now we're left looking at the wreckage of a very expensive, very organized dream. The Mechanics of the S...

Anthropic just gave us a knob for Claude's brain. For the first time, you can explicitly tell the model how much computational effort to put into a specific task. It's a subtle shift, but it changes the relationship from "send a prompt and hope for the best" to actually managing the model's reasoning process. The release includes a lot of other updates, too. The new Claude 3.5 Sonnet shows improvements in coding and agentic workflows, and the team ran the usual alignment checks to make sure it isn't behaving erratically. Most of the benchmarks look solid, showing better performance on practical knowledge work and reasoning compared to the previous version. I'm curious to see if this extra compute actually translates to better results for complex tasks, or if it's just another way to burn through API credits. We need to find out if being able to dial up the effort actually makes the model smarter, or if it just makes it slower. The mechanics ...

The idea that being available 24/7 is a sign of dedication is a lie. It’s actually a leading indicator that your architecture and your processes are broken. If you have to be constantly firefighting, you haven't built something resilient. You've just built something that requires a human sacrifice to keep running. We're being told that AI is about to 10x the productivity of the white-collar workforce. The math is pretty simple, even if the implications are messy. If these tools actually work, I should be able to produce my entire week's worth of output by Monday at noon. This creates a massive tension in how we measure value. If the work itself becomes cheap and fast, what happens to the person who is still billing by the hour or measuring success by the number of emails sent? We need to figure out if we're actually getting more efficient, or if we're just creating a new kind of digital burnout. The myth of the heroic developer The "hero develop...