I’m honestly kind of amazed that more people aren’t seeing the value, because my experience has been almost the opposite of what you’re describing.
I agree with a lot of your instincts. Shipping unreviewed code is wrong. “Validate behavior not architecture” as a blanket rule is reckless. Tests passing is not the same thing as having a system you can reason about six months later. On that we’re aligned.
Where I diverge is the conclusion that agentic coding doesn’t produce net-positive results. For me it very clearly does, but perhaps it's very situation or condition dependent?
For me, I don’t treat the agent as a junior engineer I can hand work to and walk away from. I treat it more like an extremely fast, extremely literal staff member who will happily do exactly what you asked, including the wrong thing, unless you actively steer it. I sit there and watch it work (usually have 2-3 agents working at the same time, ideally on different codebases but sometimes they overlap). I interrupt it. I redirect it. I tell it when it is about to do something dumb. I almost never write code anymore, but I am constantly making architectural calls.
Second, tooling and context quality matter enormously. I’m using Claude Code. The MCP tools I have installed make a huge different: laravel-boost, context7, and figma (which in particular feels borderline magical at converting designs into code!).
I often have to tell the agent to visit GitHub READMEs and official docs instead of letting it hallucinate “best practices”, the agent will oftentimes guess and get stack, so if it's doing that, you’ve already lost.
Third, I wonder if perhaps starting from scratch is actually harder than migrating something real. Right now I’m migrating a backend from Java to Laravel and rebuilding native apps into KMP and Compose Multiplatform. So the domain and data is real and I can validate against a previous (if buggy) implimentation). In that environment, the agent is phenomenal. It understands patterns, ports logic faithfully, flags inconsistencies, and does a frankly ridiculous amount of correct work per hour.
Does it make mistakes? Of course. But they’re few and far between, and they’re usually obvious at the architectural or semantic level, not subtle landmines buried in the code. When something is wrong, it’s wrong in a way that’s easy to spot if you’re paying attention.
That’s the part I think gets missed. If you ask the agent to design, implement, review, and validate itself, then yes, you’re going to get spaghetti with a test suite that lies to you. If instead you keep architecture and taste firmly in human hands and use the agent as an execution engine, the leverage is enormous.
My strong suspicion is that a lot of the negative experiences come from a mismatch between expectations and operating model. If you expect the agent to be autonomous, it will disappoint you. If you expect it to be an amplifier for someone who already knows what “good” looks like, it’s transformative.
So while I guess plenty of hype exists, for me at least, they hype is justified. I’m shipping way (WAY!) more, with better consistency, and with less cognitive exhaustion than ever before in my 20+ years of doing dev work.
Probably looking for a name that is much more approachable for a wide audience. Along the lines of "take me back" or playing with where/when or time/place or something like that.
They didn't acknowledge that it has gotten lazier, they've acknowledged that people have said it's gotten lazier. In fact they say it hasn't been updated since early November
> we've heard all your feedback about GPT4 getting lazier! we haven't updated the model since Nov 11th, and this certainly isn't intentional. model behavior can be unpredictable, and we're looking into fixing it
This contradicts itself each sentence and sounds as if it was written by chatgpt itself in lazy mode.
Pretty sure this comes with it's own standalone DJI remote (which is really just running android underneath). No need to sideload or connect to your phone in any way.
FWIW, ThrillSeekerVR who helped who provided some unpaid feedback on the development of this headset, also mentions in his video that he personally measured the FOV as higher than the specs list and quotes the CEO of Bigscreen as saying that he plans to "underpromise and overdeliver" on that front: https://youtu.be/mieWJ9vupqw?t=183
I don't think this works very well. I tested with a pretty popular reddit userMrPennywhistle, aka Destin from Smarter every day:
"write a reddit post about monitors in the style of reddit user MrPennywhistle"
and the response it gave me was incredibly generic and didn't sound like Destin at all:
"Hey there fellow Redditors!
I've been in the market for a new monitor lately and I have to say, it's been a real trip. There are so many options out there and it's hard to know what to choose. But after doing some research and going to a few stores to check out the options in person, I've come to a conclusion: monitors are awesome!
There are all sorts of sizes, shapes, and features to choose from, and each one has its own unique benefits. For example, some monitors are great for gaming, while others are perfect for watching movies or working on creative projects. And if you're really looking to up your setup, you can even get a curved monitor for an immersive experience.
But the most important thing to consider when buying a monitor is the resolution. A higher resolution means more pixels, which means a sharper and more detailed image. So if you're looking to upgrade your monitor, make sure to get one with a high resolution.
Overall, I'm really excited about my new monitor and I can't wait to see what it can do. Have any of you guys recently gotten a new monitor? Let me know in the comments!
I agree with a lot of your instincts. Shipping unreviewed code is wrong. “Validate behavior not architecture” as a blanket rule is reckless. Tests passing is not the same thing as having a system you can reason about six months later. On that we’re aligned.
Where I diverge is the conclusion that agentic coding doesn’t produce net-positive results. For me it very clearly does, but perhaps it's very situation or condition dependent?
For me, I don’t treat the agent as a junior engineer I can hand work to and walk away from. I treat it more like an extremely fast, extremely literal staff member who will happily do exactly what you asked, including the wrong thing, unless you actively steer it. I sit there and watch it work (usually have 2-3 agents working at the same time, ideally on different codebases but sometimes they overlap). I interrupt it. I redirect it. I tell it when it is about to do something dumb. I almost never write code anymore, but I am constantly making architectural calls.
Second, tooling and context quality matter enormously. I’m using Claude Code. The MCP tools I have installed make a huge different: laravel-boost, context7, and figma (which in particular feels borderline magical at converting designs into code!).
I often have to tell the agent to visit GitHub READMEs and official docs instead of letting it hallucinate “best practices”, the agent will oftentimes guess and get stack, so if it's doing that, you’ve already lost.
Third, I wonder if perhaps starting from scratch is actually harder than migrating something real. Right now I’m migrating a backend from Java to Laravel and rebuilding native apps into KMP and Compose Multiplatform. So the domain and data is real and I can validate against a previous (if buggy) implimentation). In that environment, the agent is phenomenal. It understands patterns, ports logic faithfully, flags inconsistencies, and does a frankly ridiculous amount of correct work per hour.
Does it make mistakes? Of course. But they’re few and far between, and they’re usually obvious at the architectural or semantic level, not subtle landmines buried in the code. When something is wrong, it’s wrong in a way that’s easy to spot if you’re paying attention.
That’s the part I think gets missed. If you ask the agent to design, implement, review, and validate itself, then yes, you’re going to get spaghetti with a test suite that lies to you. If instead you keep architecture and taste firmly in human hands and use the agent as an execution engine, the leverage is enormous.
My strong suspicion is that a lot of the negative experiences come from a mismatch between expectations and operating model. If you expect the agent to be autonomous, it will disappoint you. If you expect it to be an amplifier for someone who already knows what “good” looks like, it’s transformative.
So while I guess plenty of hype exists, for me at least, they hype is justified. I’m shipping way (WAY!) more, with better consistency, and with less cognitive exhaustion than ever before in my 20+ years of doing dev work.