Arek

2.4K posts

Arek banner
Arek

Arek

@arekhalpern

APM @AlphaSenseInc - Web Curation - opinions are my own

NYC Katılım Nisan 2021
319 Takip Edilen778 Takipçiler
Sabitlenmiş Tweet
Arek
Arek@arekhalpern·
elucide updates ✔️ backend crawler to enrich the knowledge that elucide chat will have access to, an internal crawler is table stakes. introducing elucide-crawl. to make it fun i added a bloomberg terminal style ui implementation details: - 100% python - postgreSQL database with schema-based storage - real-time metrics with 5s persistence - depth-aware crawling with configurable limits - URL deduplication and frontier management - response compression and header normalization - user agent rotation for politeness - error tracking and retry mechanisms - structured data extraction pipeline - session-based crawl management - bloomberg-style terminal UI: - real-time metrics panel - active sessions monitor - live crawl logs - command interface - sub-second UI refresh rates - CPU/memory usage tracking - response time analytics - depth-aware progress tracking
Arek@arekhalpern

elucide updates ✔️ web search seems like a simple feature to implement but it complicated because my architecture is optimized for low-latency thread posting/patching. the demo showcases elucide in action as a cloud-based webapp. its faster then chatgpt at crud. why would you want your data stored in the cloud? for long term memory which equals personalization. websearch implementation details: - Implemented handleToolEvent callback for agent mode state transitions - Added memoized shouldShowAgentSteps for temporary/permanent thread management - Enhanced stream processing to parse tool events in content messages - Optimized setIsWaitingForFirstToken for agent mode operations - Automated web search command prefixing - Implemented 1s timeout for tool event cleanup - Added thread state transition handling - Implemented temporary thread cleanup with state preservation

English
1
0
5
1.4K
Arek
Arek@arekhalpern·
@pmarca Have you seen the latest commencement speech reactions?
English
0
0
2
29
Marc Andreessen 🇺🇸
Marc Andreessen 🇺🇸@pmarca·
1. Anything that is in the world when you’re born is normal and ordinary and is just a natural part of the way the world works. 2. Anything that’s invented between when you’re fifteen and thirty-five is new and exciting and revolutionary and you can probably get a career in it. 3. Anything invented after you’re thirty-five is against the natural order of things. —Douglas Adams
Drew Pusateri@drewpusateri

Since joining OpenAI the amount of congressional staffers that've (very kindly and politely) reached out abt careers in AI/tech from offices whose Reps/Senators rail against AI/tech/infra is...notable. Tbc, there's absolutely nothing wrong with that and I'm always happy to chat and help people connect with opportunities/networking etc. I didn't agree with the electeds I worked for on everything either, but the divisions there feel a lot wider than on most issues.

English
105
278
3.7K
370.1K
Jack Zumwalt
Jack Zumwalt@jackzumwalt·
Introducing Dashboards... tomorrow. Dashboards are a collection of AI-generated widgets on a flexible canvas that maintain themselves, infinitely. You can create dashboards, edit them directly, and trust that the data is maintained for you at all times. Real-time data is critical to financial markets. When you make a visualization that you review everyday, it better maintain itself. Not only this, but we're upgrading to multi-agent orchestration with Haiku and Sonnet - indisputably the best financial reasoning models on the market. The reason I'm telling you today versus launching tomorrow is because we are raising prices, but not for our early users. Early users provide us value in many ways. Feedback, loyalty, direct contributions. We are grateful for all of them. If you join @KimptonAI in the next 24 hours, the next 3 months will only be $50/m. That price will quadruple tomorrow. Right now, we are subsidizing users in many ways to pay them for their early adoption. I want to give everyone one last wave. Be a part of our movement. We aren't going anywhere and will not stop until you have exactly what you need. And yes, I will be reaching out to you directly for feedback (so be ready!).
English
24
29
508
49.1K
signüll
signüll@signulll·
starting a consumer company is basically opting into pain as a lifestyle. you need this weird, almost contradictory stack which is taste + timing, future intuition + present execution, & culture fluency + product rigor. most ppl have like… one of these. maybe two if they’re lucky. oh the worst thing is that there’s no clean feedback loop. ain’t no tidy dashboards telling you you’re right. it’s mostly vibes, weak signals, & ridiculously delayed validation if it comes at all (kinda like being in a toxic situationship). you’re effectively betting on something that doesn’t fully exist yet, using instincts you can’t quite articulate, in a market that will happily ignore you until it suddenly doesn’t. this is one of the most asymmetric games you can play but also one of the least coherent while you’re in it. i think of it as playing a video game in hall of fame difficulty with no tutorial & half the UI missing.
TBPN@tbpn

Airbnb CEO @bchesky says more AI founders should be starting consumer businesses. "I'm on the board of Y Combinator. 87% of companies are enterprise companies per batch." "Enterprise is awesome... but the biggest prize is consumer. That is what's going to reach daily life for billions of people." "Think about all the little parts of daily life that are kind of annoying. Pay attention to whoever's in your life and ask: 'How could their daily life be a little bit easier?'" From his appearance on the show in January.

English
78
88
1.3K
192.9K
Arek
Arek@arekhalpern·
Main issue with Claude - lack of discernment
Arek tweet media
English
0
0
0
71
Arek
Arek@arekhalpern·
@karpathy funny that human tendency to forget things sometimes is a feature not a bug. unsure how we’ll emulate this in llms
English
0
0
1
55
Andrej Karpathy
Andrej Karpathy@karpathy·
One common issue with personalization in all LLMs is how distracting memory seems to be for the models. A single question from 2 months ago about some topic can keep coming up as some kind of a deep interest of mine with undue mentions in perpetuity. Some kind of trying too hard.
English
1.8K
1.1K
21.2K
2.7M
Arek
Arek@arekhalpern·
Children chase robot in New York City
English
0
0
1
49
Arek
Arek@arekhalpern·
@DavidSHolz @bdmarotta @midjourney Couldn’t you just compete directly with fal/replicate. I know you are different companies but adjacent nonetheless
English
0
0
0
65
David
David@DavidSHolz·
@bdmarotta @midjourney its not clear an API will drive much revenue right now, no one is making 'enough images' for a pay-per-image biz model to work. there might be a 'higher subscription' play that works for professionals and small biz though, but still not obvious it is a huge boon
English
28
2
79
7.2K
John Haslbauer
John Haslbauer@PGATout·
The Wizards trading for Trae Young + AD and benching them both for the rest of the season is a blatant sign that the NBA Lottery system has failed to stop tanking. I came up with a solution to fix it:
John Haslbauer tweet mediaJohn Haslbauer tweet mediaJohn Haslbauer tweet mediaJohn Haslbauer tweet media
English
110
159
2.3K
1.2M
Hassan Hayat 🔥
Hassan Hayat 🔥@TheSeaMouse·
btw, this is the same trick people have been using lately to get agents to navigate repos faster
Nishkarsh@contextkingceo

We've raised $6.5M to kill vector databases. Every system today retrieves context the same way: vector search that stores everything as flat embeddings and returns whatever "feels" closest. Similar, sure. Relevant? Almost never. Embeddings can’t tell a Q3 renewal clause from a Q1 termination notice if the language is close enough. A friend of mine asked his AI about a contract last week, and it returned a detailed, perfectly crafted answer pulled from a completely different client’s file. Once you’re dealing with 10M+ documents, these mix-ups happen all the time. VectorDB accuracy goes to shit. We built @hydra_db for exactly this. HydraDB builds an ontology-first context graph over your data, maps relationships between entities, understands the 'why' behind documents, and tracks how information evolves over time. So when you ask about 'Apple,' it knows you mean the company you're serving as a customer. Not the fruit. Even when a vector DB's similarity score says 0.94. More below ⬇️

English
5
1
61
20.5K
Nishkarsh
Nishkarsh@contextkingceo·
We've raised $6.5M to kill vector databases. Every system today retrieves context the same way: vector search that stores everything as flat embeddings and returns whatever "feels" closest. Similar, sure. Relevant? Almost never. Embeddings can’t tell a Q3 renewal clause from a Q1 termination notice if the language is close enough. A friend of mine asked his AI about a contract last week, and it returned a detailed, perfectly crafted answer pulled from a completely different client’s file. Once you’re dealing with 10M+ documents, these mix-ups happen all the time. VectorDB accuracy goes to shit. We built @hydra_db for exactly this. HydraDB builds an ontology-first context graph over your data, maps relationships between entities, understands the 'why' behind documents, and tracks how information evolves over time. So when you ask about 'Apple,' it knows you mean the company you're serving as a customer. Not the fruit. Even when a vector DB's similarity score says 0.94. More below ⬇️
English
620
638
6K
3.9M
Arek
Arek@arekhalpern·
kinda crazy
Nik Cubrilovic@dir

@pmddomingos gmail, adsense, aws, slack, fb messenger, safari, github copilot, instagram and the site you're posting this on were all started as side projects

English
0
0
0
123
Arek
Arek@arekhalpern·
What happened to n8n? The original clawdbot
Arek tweet media
English
0
0
1
91

Keşfet

@pmarca @netcapgirl @XP_Ehsaan @jackzumwalt @KimptonAI @signulll @karpathy @DavidSHolz