motoboi

6 ms·

31.
by motoboi 4mo ago
Not sure, but in my experience, instead of asking for code, i'm asking for solutions and providing a kubectl configured to reach my cluster and az monitor command to read the logs and telemetry. A typical session is the agent establish
32.
by motoboi 4mo ago
There’s nothing much new about the architecture. The real gains come from the usage traces. It turns out that having a text based interface for a text-trained model creates a very nice feedback loop. Right now as we speak, people are genera
33.
by motoboi 5mo ago
Very nice. But nowhere to be seen in my model list on github copilot enterprise ai settings? I suppose it's still rolling out. The "rolling out to github copilot" is verbatim on the blog post, not my words. On the other hand,
34.
by motoboi 5mo ago
Unless you are token rich, you'll have to find a way pretty soon. For tasks (like kubernetes, linux, reports, database exploration and such) I use GLM5.1. Faster is actually smarter in those cases. And much cheaper too. Opus 4.8 is for
35.
by motoboi 5mo ago
To understand microsoft IA problems right now, observe that NONE of the models announced are available for use even in the microsoft foundry, which is the place were you add models to your account. I understand github copilot rollout takes
36.
by motoboi 5mo ago
For me the greatest difference is that superset is terminal-centric while conductor is chat-centric. GUIs slow you down, in my opinion. But having the nice visual diff is something we can't really do well in TUIs, so very welcome. So,
37.
by motoboi 5mo ago
I strongly believe you don’t need to call another model for that. The same model can do result fine. Just not as part of the same context. I mean that if you ask codex on gpt 5.5 to submit to a plan reviewer subagent that uses gpt5.5, this
38.
by motoboi 5mo ago
The microbiota is passed from mother to son on birth, not totally from the environment. What we currently don’t understand is why for some people they never got them (we have techniques to transport the biota from the mother during birth fo
39.
by motoboi 6mo ago
Not apples for apples. Before: - Opus 4.6 each premium request is 3 premium requests After: - Opus 4.6 each dollar spent is 27 dollars in copilot AI Credits. Given that you'll receive 19 dollars of AI Credits in Business plan, that mea
40.
by motoboi 6mo ago
Wait until you discover how many wrong labeled images in imagenet and that it still kickstarted the deeplearning revolution.
41.
by motoboi 6mo ago
It’s saying that 16% of the problems have well, problems.
42.
by motoboi 6mo ago
Please next time start with azure foundry lol thanks!
43.
by motoboi 6mo ago
This is how anthropic captured the code agent so fast. You need training data, users are giving it to you. Being a terminal application, all interaction is trainable signal (unlike, say, cursor, which is an IDE and let users freely explore,
44.
by motoboi 6mo ago
Well, the model that a lot of people have been given access to and are reporting about on twitter?
45.
by motoboi 6mo ago
With the previous generation? Yes. With 10T mythos-level models? Not even close.
46.
by motoboi 6mo ago
Don't. thinking right now is just text. Chain of though, but just regular tokens and text being output by the model.
47.
by motoboi 6mo ago
Caveman hurt model performance. If you need a dumber model with less token output, just use sonnet-4-6 or other non-reasoning model.
48.
by motoboi 6mo ago
The reasoning is the secret sauce. They don't output that. But to let you have some feedback about what is going on, they pass this reasoning through another model that generates a human friendly summary (that actively destroys the sig
49.
by motoboi 6mo ago
People say we don’t build cathedrals anymore. But here it is: JVM is a modern cathedral.
50.
by motoboi 7mo ago
I believe what is missed here is that the brain assimilate things outside of it (not in the physical sense of course). Use a hammer for some time and the brain start to dedicate some networks to simulate the hammer internally and to integra
51.
by motoboi 7mo ago
> it was too easy to put YouTube running That is what is killing your focus and giving ADHD-like symptoms (supposing obviously that you don’t actually have it). Never ever put something to play on background while you do something. That
52.
by motoboi 7mo ago
Im sorry. You are thinking in terms of one time interactions. I’m thinking about the next step in the interaction. To understand my point, think about a prompt to tell the model “here is a very difficult code problem, answer in a single wor
53.
by motoboi 7mo ago
The reasoning part is not diferente from the part that goes in answer. It’s just that the model is trained to do some magical text generation with back and forth. But when it’s writing the answer part of it, each word is part of its context
54.
by motoboi 7mo ago
The output is part of context. The model reason but also output tokens. Force it to respond in an unfamiliar format and the next token will veer more and more from the training distribution, rendering the model less smart/useful.
55.
by motoboi 7mo ago
If they want to reduce token cost, just use a smaller model instead of dumbing down a more expensive.
56.
by motoboi 7mo ago
Things like this make me sad because they make obvious that most people don’t understand a bit about how LLM work. The “answer before reasoning” is a good evidence for it. It misses the most fundamental concept of tranaformers: the are auto
57.
by motoboi 7mo ago
I insist on the text generator nature of the thing. It’s just that we built harnesses to activate on certain sequences of text. Think of it as three people in a room. One (the director), says: you, with the red shirt, you are now a plane co
58.
by motoboi 7mo ago
The whole “chat with an AI” paradigm is the culprit here. Priming people to think they are actually having a conversation with something that has a mind model. It’s just a text generator that generates plausible text for this role play. But
59.
by motoboi 7mo ago
Working on TRIPA, a internal pack format to save huge amount of small images to cloud storage without dying from the write costs.
60.
by motoboi 8mo ago
Im planning a change that will save 20k a month of storage. I absolutely could come up with the details and implementation by myself, but that would certainly take a lot of back and forth, probably a month or two. I’m an api user of Claude

More ›