veunes

12 ms·

211.
by veunes 10mo ago
If a business model only works as long as customers don't notice the real price, then the model itself is fundamentally broken
212.
by veunes 10mo ago
What's interesting is how uneven the landscape is across states
213.
by veunes 10mo ago
When the goal isn't to run a good store but to extract value as fast as possible, all the classic PE patterns show up
214.
by veunes 10mo ago
The saddest part is the feedback loop
215.
by veunes 11mo ago
This is the inevitable evil of the man in the middle. OpenRouter by definition decrypts your traffic to route it to the provider (OpenAI, Anthropic). Technically, they can read everything The problem is that for the Enterprise segment, this
216.
by veunes 11mo ago
I'm not surprised. Roleplay means endless sessions with huge context (character history, world, previous dialogues). On commercial APIs (OpenAI/Anthropic), that long-context costs a fortune. On OpenRouter, many OSS models, especia
217.
by veunes 11mo ago
Yeah, using an API aggregator to run a 7B model is economically strange if you have even a consumer GPU. OpenRouter captures the cream of complex requests (Claude 3.5, o1) that you can't run at home. But even for local hosting, medium
218.
by veunes 11mo ago
The 4x growth in prompt length is a fundamental shift. We've quickly moved from "Q&A" mode to "upload full context and analyze" mode. This completely changes infrastructure requirements: KV-caching becomes a nec
219.
by veunes 11mo ago
Timing is everything here. The swiss army knife approach only loses when tasks stop changing. Intel suffered when workloads like web and mobile stabilized In AI, we're still in the explosion phase. If you build the perfect ASIC for Tra
220.
by veunes 11mo ago
That engine had character
221.
by veunes 11mo ago
What's tricky is that even tiny improvements in fuel economy or emissions can justify a redesign when you're building at scale
222.
by veunes 11mo ago
Plenty of engines hit the 20-year mark, but not many do it while still powering everything from family sedans to track-ready Type Rs
223.
by veunes 11mo ago
I think the article's broader point still stands: Honda built a platform with enough foresight and flexibility that it could be continuously refined rather than scrapped and replaced every few years
224.
by veunes 11mo ago
What's impressive isn't just the longevity, but how gracefully it's evolved over two decades while still feeling relevant in today's turbocharged, emissions-strangled landscape
225.
by veunes 11mo ago
The own shovels for own mines strategy has a hidden downside: isolation. NVIDIA sells shovels to everyone - OpenAI, Meta, xAI, Microsoft - and gets feedback from the entire market. They see where the industry is heading faster than Google,
226.
by veunes 11mo ago
OCS is indeed an engineering marvel, but look at NVIDIA's NVL72. They took a different path: instead of flexible optics, they used the brute force of copper, turning an entire rack into one giant GPU with unified memory. Google is solv
227.
by veunes 11mo ago
The problem is that "hardware-agnostic PyTorch" is a myth, much like Java's "write once, run anywhere". At the high level (API), the code looks the same, but as soon as you start optimizing for performance, you inev
228.
by veunes 11mo ago
The funniest thing about this story is that NVIDIA has essentially become a TPU company. Look at the Hopper and Blackwell architectures: Tensor Cores are taking up more space, the Transformer Engine has appeared, and NVLink has started to l
229.
by veunes 11mo ago
The idea is interesting, but it's still operating within the content analysis paradigm. As soon as entropy-based detectors become popular, the next generation of LLMs will be specifically fine-tuned to generate higher-entropy text to e
230.
by veunes 11mo ago
The CEOs obstinacy comes from simple economics: the cost of producing content with AI is trending toward zero, which allows for scaling content farms to unprecedented sizes. It's a constant race for attention, so the goal is no longer
231.
by veunes 11mo ago
This is an inevitable arms race. The slop generators will constantly improve to fool the detector, and the detector will have to train on their new tricks The problem is that pure content-based analysis (at the text or image artifact level)
232.
by veunes 11mo ago
Windows is definitely not the best place for this kind of thoughtless integration, because the digital lives of billions of people around the world depend on it
233.
by veunes 11mo ago
They're trying to bolt an unpredictable language engine onto an OS whose security model is based on total predictability. It's basically a rootkit that you consent to installing. No audit logs will help after the agent has already
234.
by veunes 11mo ago
At some point, the trade-offs just shift. Manual memory control feels empowering until the complexity piles up, and suddenly GC looks like a smart compromise
235.
by veunes 11mo ago
It's amazing how often symmetry shows up as both a design aesthetic and a practical tool for managing complexity
236.
by veunes 11mo ago
It's not just about what it can't do (like handle cycles on its own), but about what it enables in certain contexts
237.
by veunes 11mo ago
Maybe not a full philosophical dual, but still a very practical one in this context
238.
by veunes 11mo ago
It's always fun when a concept you learned years ago suddenly snaps into place for a totally different problem domain
239.
by veunes 11mo ago
Careful now, you're gonna jinx it
240.
by veunes 11mo ago
It doesn't have to dethrone anyone. If SteamOS-powered boxes start eating into the "enthusiast console" nich that alone could force Microsoft and Sony to rethink their current lock-in-heavy strategies

More ›