firasd

8 ms·

91.
by firasd 2mo ago
Seems like Instagram (and TikTok) are a big part of the story here. A lot of these 'buy my hat with a funny slogan' or 'buy my lotion' stores on Shopify, Etsy etc are highly dependent on direct-response ads / perfor
92.
by firasd 2mo ago
Connecting 'this account belongs to you' to 'this account did these activities' seems to rely on a chain of high-integrity (in the sense of both ethics and precision) investigative work that clearly doesn't happen b
93.
by firasd 2mo ago
Interesting. It reminds me of matching code ranges with and without whitespace.. the technique is useful in many contexts like source maps, minifiers, diff tools
94.
by firasd 2mo ago
Dario has like three 'paranoias' / strong-motivating-concerns 1) LLMs turning into Skynet 2) China as geopolitical competitor 3) Claude being 'distilled' by competitors (this has led Anthropic to cut service to vari
95.
by firasd 2mo ago
Interesting... yeah if you think of Docker as an easy way to setup environments that's one thing but if you are intending airtight isolation so processes inside Docker can't "escape" most conventional use cases / di
96.
Blade Runner 2099 first look from showrunner Silka Luisa(ew.com)
3 points by firasd 2mo ago|0 comments
97.
Data loss in Claude Code and OpenAI Codex: when AI deletes user files(firasd.substack.com)
2 points by firasd 2mo ago|0 comments
98.
by firasd 2mo ago
Would be interesting to have a non-interactive mode so you can ask your AI to look for a command too.. but I guess if it's sqlite underneath then it's already de facto possible (and the the cltr+r may be usable from harnesses too
99.
Wreckage of 1952 Pan Am plane crash found near Puerto Rico(theguardian.com)
3 points by firasd 3mo ago|0 comments
100.
by firasd 3mo ago
I've always thought that extensive throat-clearing and prefixing the Treaties of Westphalia-length instructions into the context window was unnecessarily baroque when you can just talk to the agent. I guess part of it is also that I do
101.
by firasd 3mo ago
Very interesting. The past few days' trend is so common in small talk and general weather discourse ("it was warmer last week") but underrated in weather apps. It's helpful that Open-Meteo provides that data This output
102.
by firasd 3mo ago
Not really though right? As of early June 2026, Opus 4.8 in fast mode cost $50/M output tokens and Opus 4.6 & 4.7 cost $150/M output tokens in fast mode How can supply and demand explain the price drop? Was it cheaper to serve
103.
by firasd 3mo ago
One of the strangest things in the AI industry is 'tokenomics'. It's not very clear why using GPT-4 in early 2023 was so expensive and then six months later 20 bucks could get you a fair amount of GPT-4 inference. This patter
104.
by firasd 3mo ago
I think it depends on the context too. Dale Carnegie (in his famous Win Friends.. book) mentioned how it's annoying to receive letters that say Dictated But Not Read. So people don't want to read (for example) a twitter reply or e
105.
by firasd 3mo ago
This is like the late 80s/early 90s cyberspace idea... Imagine you're walking through a 3D ramp and all these buttons are yelling at you from all over the atmosphere... Login here! Start here! Contact us...
106.
by firasd 3mo ago
I think this nostalgia is useful for reminding us there are aesthetic forms beyond what Jony Ive et al happened to design and became the dominant form factor. The 2000s explosion of MP3 players had a lot of little ideas like the iRiver Clix
107.
by firasd 3mo ago
Grounding as in web search? I think this Omniscience benchmark would not include access to a search tool cause otherwise it becomes kinda meaningless
108.
by firasd 3mo ago
Hi all, I figured since it's been 2.5 years since ChatGPT launched and the labs haven't seen fit to let their default apps have a time tool, here's one: Live MCP endpoint: https://mcpclock.firasd.workers.dev/m
109.
MCP Clock: live public hosted tool for Claude, Codex, any MCP Client(github.com)
2 points by firasd 3mo ago|1 comments
110.
by firasd 3mo ago
The idea in all the examples in this page seems to be: write a string with some code and then write some more code that processes the string in a way that when the string is finally output it contains the variable assignment (of the string)
111.
by firasd 3mo ago
Very interesting that one of the components is "AA-Omniscience Index" AA-Omniscience Index (higher is better) measures knowledge reliability and hallucination. It rewards correct answers, penalizes hallucinations, and has no penal
112.
by firasd 3mo ago
Very interesting. Here in India people were very concerned about potential cooking gas shortages (LPG) when the disruptions began which is also a good example of usually-overlooked dependencies on the the Strait
113.
by firasd 3mo ago
I think one way to look at this is that--why are Meta and Google so successful? Because they are basically an index fund on the world economy right. If you want to sell soup in Jakarta you advertise on Instagram. If you want to sell paint i
114.
Billy Batts That 'Goodfellas' Didn't Tell(allthatsinteresting.com)
3 points by firasd 3mo ago|0 comments
115.
by firasd 3mo ago
I think there is a fundamental issue here of what building software even means If you think you can just assign Github tickets to AI agents and go drink daiquiris on the beach I think you'll find that you end up with more and more to
116.
by firasd 3mo ago
I have been seeing a few tweets about AI agents deleting data and I figured there is a pattern in the examples I've seen. From the article: Example 1: $HOME didn’t address the directory assumed; PowerShell’s -Include parameter didn’t w
117.
Accidental data loss in Claude Code and OpenAI Codex: when AI deletes user files(firasd.substack.com)
2 points by firasd 3mo ago|1 comments
118.
by firasd 3mo ago
Have you ever tried analyzing a CSV in an LLM? Maybe then you'd get what I'm saying Heck for a lot of questions the LLM won't even write Pandas and just try to eyeball the arithmetic and be like "Yeah March 3 to March 7
119.
by firasd 3mo ago
An absurd example is how harnesses like Codex, Claude code have no 'cut/paste this range of chars' affordance so if you want to move a block of code the LLM galaxy brain rewrites it in token space in the new file. It ~can~ be
120.
by firasd 3mo ago
Yeah I guess the dichotomy between 'tools' and 'code mode' isn't really as clear especially if the tools themselves are intelligent and truncate long results and have start_range, end_range type arguments. So the gi

More ›