ttul

10 ms·

151.
by ttul 7mo ago
This is so true. I don’t like Telegram for a host of reasons, but the bot architecture is second to none. Try creating a bot in Slack. You’ll pull your hair out for hours. Same goes for Discord. Utter nightmare. Telegram? You send a DM and
152.
by ttul 7mo ago
There’s always room for improvement…
153.
by ttul 7mo ago
Case in point: my parents. Built a house in 1988 and they still live there. Two people in 3500 square feet. Four bathrooms and five bedrooms. Meanwhile, you need a family income of 3x the median to rent a townhouse 1/3rd the size nearb
154.
by ttul 7mo ago
Starlink seems to provision capacity by locking your service to one address at a time; presumably, this means they have enough capacity for the customers in each physical area. By contrast, mobile networks have to contend with highly mobile
155.
by ttul 7mo ago
I bought the original AirPods Max when they came out a few years ago. They hurt my ears if I wear them for more than an hour or so and the software is garbage - constantly losing one channel or just failing to connect at all - to Apple devi
156.
by ttul 7mo ago
The issue is that mobile is easily overloaded if those around you are also failing over onto it. There are only so many channels available per sector. In my experience, when one of the two incumbent carriers in my area goes down, mobile is
157.
by ttul 7mo ago
It’s the business cycle, mostly. During the pandemic, low interest rates drove a boom in risk investing that flowed downhill into tech company balance sheets. Of course everyone used the money to hire lots of developers and engineers - prob
158.
by ttul 7mo ago
em-dash -> permaban?
159.
by ttul 7mo ago
I’m not quite as old as you, but I am old enough to know what a COM component is and to have ready the Byte Magazine article that likely described this ancient stone tablet tech. Codex has me absolutely stoked again. I can finally have fun
160.
by ttul 7mo ago
Next time you’re using your favorite LLM as a therapist, try editing your previous input and getting it to regenerate its response. It’s a humbling experience to see your trusted “therapist” shift from one perspective or piece of advice to
161.
by ttul 7mo ago
Cool approach. So basically the part that needs to be realtime - the voice that speaks back to you - can be a bit dumb so long as the slower-moving genius behind the curtain is making the right things happen.
162.
by ttul 7mo ago
What gives you the sense that the piece was written by an LLM? I would agree that the diagrams have some of the artifacts common in Nano Banana output, but what tips you off about the text?
163.
by ttul 8mo ago
I don’t make them do that. They do it because they want to be ready to help in the seconds that matter if someone has an overdose. This is the sad reality they inhabit.
164.
by ttul 8mo ago
I stand corrected here. The design is "fully digital". They are not using analog trickery here. They are most likely just using a clever collection of very tiny transistors hooked up to two wire layers such that one _digital_ tran
165.
by ttul 8mo ago
I have several naloxone kits in my house and my kids carry them in their backpacks. I'm pro harm-reduction. With that in mind, what I'm "actually trying to say" is that a) any time we can make a medication less harmful,
166.
by ttul 8mo ago
Didn't AirBnB famously spam people in the Bay Area as a "guerilla tactic" to build the business in its early days? This kind of fast and loose behaviour seems standard.
167.
by ttul 8mo ago
Well, if it doesn't suppress breathing dangerously, and yet works on "pain" effectively, it would be the most widely abused drug on the planet...
168.
by ttul 8mo ago
Yeah this is totally wrong. Taalas says their design is fully digital.
169.
by ttul 8mo ago
LoL. Some day soon, we will all need to carry proof of humanness and have our English Literature transcripts from university to evidence that we did, actually, learn to write at some point.
170.
by ttul 8mo ago
(I am not a bot, but I definitely use LLMs to help me conduct research and write well-informed comments here)
171.
by ttul 8mo ago
Which user is an AI bot?
172.
by ttul 8mo ago
Trading on non-public information in prediction markets is illegal (in the United States) if the information was obtained through fraud, deception, a breach of trust—such as compromising a position of privilege—or from a confidential govern
173.
by ttul 8mo ago
Regarding models, I’ve found that going with OpenRouter’s `auto` model works well enough, choosing the powerful models when they seem to be needed, and falling back on cheaper ones for other queries. But, it’s still expensive… Depending on
174.
by ttul 8mo ago
The NextPlatform article hints at their approach: “We have got this scheme for the mask ROM recall fabric – the hard-wired part – where we can store four bits away and do the multiply related to it – everything – with a SINGLE TRANSISTOR. S
175.
by ttul 8mo ago
The ZC PR experience is hard core. Their PR template asks for a lot of details related to security and correctness - and they check it all before merging. I submitted a convenience script that gets ZC rolling in a container with one line. P
176.
by ttul 8mo ago
The ZeroClaw team is focusing their efforts on correctness and security by design. Observability is not yet there but the project is moving very rapidly. Their approach, I believe, is right for the long term.
177.
by ttul 8mo ago
Having dabbled in VLSI in the early-2010s, half the battle is getting a manufacturing slot with TSMC. It’s a dark art with secret handshakes. This demonstrator chip is an enormous accomplishment.
178.
by ttul 8mo ago
“Some described sending a “quick last prompt” right before leaving their desk so that the AI could work while they stepped away.” This, I can relate to. Also: I feel like I need a second monitor.
179.
by ttul 8mo ago
What I’m noticing, overall: I’ve never cut so much code in my life. I’ve become a coding monster with one of those dark green GitHub profiles ever since 5.3-Codex gave me the confidence to load in a ridiculous number of tasks every day and
180.
by ttul 8mo ago
The game theory here is that either OpenAI acquires this thing now, or someone else will. It doesn't matter whether they could replicate it. All of the major players can and probably will replicate OpenClaw in their own way and make th

More ›