i code @Qiro_Finance | 🏆 8x Web3 Hackathons | 🐧i use arch, btw | Cooking in Web3xAI
trenches rn
Joined July 2023
- Tweets540
- Following1.1K
- Followers746
- Likes611
most of the current memory systems out there, always excel at specific things, and not the others
i think @supermemory is the only memory system that seems to be handling every inputs/question types to a good extend
today for the first time ever i found a model, opus 5.5 at max effort, responded me saying my suggestion is shit and it would rather do it another way
am having mixed feelings about this xD
trying to build a memory system that is good at answering any questions (temporal facts, changing facts, anything), bringing in multiple good things i read from multiple memory systems out there together
only benchmarks gonna answer if this gonna break or survive xD
well, it looks like they actually pacing ai this time fr xD
neither openai nor anthropic has released a model in a while
A new way to measure how AI agents handle the routine analyses of wet-lab science—BioBasicsBench.
Developed with Ligo, a biotech company focused on enzyme design, the benchmark draws on 100 tasks from real experiments to test whether agents can interpret raw lab data, identify errors, and recognize when the evidence doesn’t support a conclusion.
Accelerating biological research starts with getting the everyday work right. BioBasicsBench shows where models need to improve, and helps us track what those gains mean for scientific progress, biosafety, and biosecurity.
openai.com/index/measuring-h…
well, looks like we're one step closer to quantum computers
AI is turning science fiction into reality sooner than ever. xD
Today Physical Superintelligence PBC is announcing a proof of the Aaronson-Ambainis conjecture, one of the central open problems in quantum complexity, with major implications for quantum computing speedup.
psi.inc/research/aaronson-am…
Thread 👇
been a huge fan of @MiniMax_AI, ever since minimax m2.7, wasn't expecting this price bump
had so much with minimax models, esp m3.1 flash was insane, ig time to say goodbye 😭
today read a bunch of engineering blog posts from big ai companies cursor, openai, replit etc
the amount of times i saw temporal.io being used in their architecture was insane lol
its truly goat'd and underrated one fr
PLS TELL ME THIS IS AI
MEDICAL ROBOT RUNS FROM POLICE RIGHT IN THE CITY CENTER
Robots are built to follow commands. This one turned and ran on its own.
This isn't staged. This isn't a demo.
Cameras caught it live. A police car and an ambulance were already on scene. The robot in a blue medical gown bolted through the pedestrian zone, weaving straight through café tables.
It dodged between chairs. It moved around people without slowing down. It never lost its balance.
Officers on foot couldn't keep pace. Step speed, balance, obstacle response, all running in real time. Nobody scripted this scenario in advance.
Autonomy isn't a glitch in the code. It's the basic wiring of a body that finally learned to move on its own.
Readers added context they thought people might want to know
Videos of a medical robot running from police are typically AI generated. No credible news outlets or authorities have reported any such real incident.
latestly.com/technology/did… annahar.com/factcheck/3452…
capGoblin retweeted
“I believe AI will cure cancer and save millions of lives…. I’m calling for a slowdown in AI development.”
We Must Pace the Frontier: I’ve written a new essay on why the AI industry should slow down, with a three-part plan for doing so.
Anthropic is unilaterally committing to the first of these steps. We’ll provide third-party evaluators with permanent, employee-level access to our systems, so that they can verify adherence to our safety measures, report on incidents, and assess models’ alignment during training.
You can read the full post here: darioamodei.com/post/we-must…
feels weird, to see this collab
We Must Pace the Frontier: I’ve written a new essay on why the AI industry should slow down, with a three-part plan for doing so.
Anthropic is unilaterally committing to the first of these steps. We’ll provide third-party evaluators with permanent, employee-level access to our systems, so that they can verify adherence to our safety measures, report on incidents, and assess models’ alignment during training.
You can read the full post here: darioamodei.com/post/we-must…
“I believe AI will cure cancer and save millions of lives…. I’m calling for a slowdown in AI development.”
We Must Pace the Frontier: I’ve written a new essay on why the AI industry should slow down, with a three-part plan for doing so.
Anthropic is unilaterally committing to the first of these steps. We’ll provide third-party evaluators with permanent, employee-level access to our systems, so that they can verify adherence to our safety measures, report on incidents, and assess models’ alignment during training.
You can read the full post here: darioamodei.com/post/we-must…
capGoblin retweeted
1/2
Qiro is bringing short-term private credit to @ConcreteXYZ’s USD1 RWA Vault.
Built with @ConcreteXYZ, the vault gives eligible USD1 investors access to a diversified portfolio of real-world opportunities through a single vault.
1/ Inside the USD1 RWA Vault: Qiro.
@Qiro_Finance is a private credit curator building short-term yield products for institutional stablecoin investors.
capGoblin retweeted
1/x
We are glad to announce a strategic partnership with @infiniFi
Qiro will act as a strategic underwriting and risk-monitoring partner to infiniFi, evaluating up to $100M in private credit opportunities over the next 12 to 15 months.
We’ve already underwritten $29M+ across New Silver and mGLOBAL.
More here: qiro.substack.com/p/infinifi…
🧵
honestly i'd be happy, can finally code by hand, it will be fun back again. would have less work, and more time.
This quoted post is unavailable.
most people's use cases dont need agents tbh
if you can define each step clearly, you dont need an agent harness
langgraph just works
