mmaunder

12 ms·

151.
by mmaunder 10mo ago
Used the hell out of Gemini 3 Flash with some 3 Pro thrown in for the past 3 hours on CUDA/Rust/FFT code that is performance critical, and now have a gemini flavored cocaine hangover and have gone crawling back to Codex GPT 5.2 xh
152.
by mmaunder 10mo ago
It's not funny when I have to explain the joke.
153.
by mmaunder 10mo ago
Agreed. It's ridiculous.
154.
by mmaunder 10mo ago
Yeah the only thing standing in Google's way is Google. And it's the easy stuff, like sensible billing models, easy to use docs and consoles that make sense and don't require 20 hours to learn/navigate, and then just the
155.
by mmaunder 10mo ago
I think about what would be most terrifying to Anthropic and OpenAI i.e. The absolute scariest thing that Google could do. I think this is it: Release low latency, low priced models with high cognitive performance and big context window, es
156.
by mmaunder 10mo ago
Thanks, having it walk a hardcore SDR signal chain right now --- oh damn it just finished. The blog post makes it clear this isn't just some 'lite' model - you get low latency and cognitive performance. really appreciate you
157.
by mmaunder 10mo ago
"...trust from other large, imporant [sic] third parties which in turn has given Waterfox users access to protected streaming services via Widevine." The black box objection disqualifies Widevine.
158.
What happens when the coding becomes the least interesting part of the work(simonwillison.net)
6 points by mmaunder 10mo ago|4 comments
159.
by mmaunder 10mo ago
I'm finding I can put my ops background fully to work, designing far more complex and performant architectures that require big lifts, without worrying about how much my fingers are going to hurt and that it'll take 8 months to ev
160.
by mmaunder 10mo ago
Solid approach. Don’t be shy about writing long prompts. We call that context engineering. The more you populate that context window with applicable knowledge and what exactly you want, the better the results. Also, having the model code an
161.
by mmaunder 10mo ago
That's like telling a pig to become a pork producer.
162.
by mmaunder 10mo ago
Yeah it's a weird mix of issues with the backend model and issues with the CLI client and its prompts. What makes it hard for them is the teams aren't talking to each other. The LLM team throws the API over the wall with a note sa
163.
by mmaunder 10mo ago
My name is Mark Maunder. Not the fisheries expert. The other one when you google me. I’m 51 and as skeptical as you when it comes to tech. I’m the CTO of a well known cybersecurity company and merely a user of AI. Since you critiqued my pos
164.
by mmaunder 10mo ago
Yeah I think a lot of us are taking knowing how LLMs work for granted. I did the fast.ai course a while back and then went off and played with VLLM and various LLMs optimizing execution, tweaking params etc. Then moved on and started being
165.
by mmaunder 10mo ago
Same. Also got my attention re ARC-AGI-2. That's meaningful. And a HUGE leap.
166.
by mmaunder 10mo ago
Weirdly, the blog announcement completely omits the actual new context window size which is 400,000: https://platform.openai.com/docs/models/gpt-5.2 Can I just say !!!!!!!! Hell yeah! Blog post indicates it's
167.
by mmaunder 10mo ago
Yeah I eventually noped out as I said in another comment and am charging hard with Codex and am so happy about 5.2!!
168.
by mmaunder 10mo ago
Then you haven't used Gemini CLI with Gemini 3 hard enough. It's a genius psychopath. The raw IQ that Gemini has is incredible. Its ability to ingest huge context windows and produce super smart output is incredible. But the bias
169.
by mmaunder 10mo ago
Thanks, that's very kind.
170.
by mmaunder 10mo ago
It's exactly what existing manufacturers are doing. What they're calling "legacy". Here, watch this which goes into detail on existing superalloy tech used in todays engines: https://youtu.be/QtxVdC7pBQM
171.
by mmaunder 10mo ago
Yeah 100% on the same page. I wrote a detailed comment here but I left out my intuition around supersonic being less efficient. Thanks for weighing in. TBO is a huge factor and the post relies on the readers naivete around that and other fa
172.
by mmaunder 10mo ago
I want these guys to succeed because they're a mile from my hangar at KAPA, but I guess key questions and points are: Industry is already using single crystal blades and superalloys, so the diagram labeling those is eye candy. No hot-d
173.
by mmaunder 10mo ago
I feel like Codex is the middle ground. You can define a project, break it into bite sized chunks, but still lift a reasonable amount. Claude with Opus 4.5 right now chews up context at an eye watering rate. It's really unfortunate bec
174.
by mmaunder 10mo ago
Big projects that are hard. Rust and CUDA for example. Particularly where you have many files that need to be fully read in for context, before planning or implementing something. Also very hard algorithmic problems. Also bugs that Claude C
175.
by mmaunder 10mo ago
No it's not. Gemini 3 is unbelievable when it works. Faster. Bigger context window. Higher intelligence as proven on ARC-AGI-2. Please skip the all-lower-case-troll comments and add actual data to the conversation.
176.
Is Gemini 3 with Gemini CLI having issues?
1 points by mmaunder 10mo ago|7 comments
177.
by mmaunder 10mo ago
This is why TDD is how you want to do AI dev. The more tests and test gates, the better. Include profiling in your standard run. Add telemetry like it’s going out of fashion. Teach it how to use the tools in AGENTS.md. And watch the output.
178.
by mmaunder 11mo ago
Thanks for doing this Mitchell, if you're on here, and for contributing to open source. Donated.
179.
by mmaunder 11mo ago
Anger is a mind killer. Build software out of love. Love for engineering, innovation, creation, and love of working with people who feel the same way.
180.
by mmaunder 11mo ago
Thanks!

More ›