Alex Morisse, PhD
@silkspaceships
Followers
391
Following
923
Media
1
Statuses
116
Founder https://t.co/xP47mznIEA, Head of AI, https://t.co/5ouMBfFvJP AI Product Eng, @LOUIE_AI The God of Figment's Imagination. UCC 1-308
California
Joined October 2010
RT @lmeyerov: Excited for @silkspaceships to share one of our AI team's biggest focuses for awhile now, including a SOTA result for #sec….
0
1
0
We’ve been wiring serious backend muscle with contextual front-end flow to handle real industry-scale data + chaos. Watch @lmeyerov go full cyber-detective mode in Vibes Investigating—red-teaming never looked this smooth.
I'm increasingly doing something we're calling "Vibes Investigating" in the @louie_ai team (ht @silkspaceships ). This rethinks @karpathy’s "Vibes Coding", now for data - think operations, incident response, fraud, SRE, logs, data science. I make requests, see results, adjust,
1
2
8
Anyone else noticed that gpt-4o seems incredibly lazy lately inside their agentic applications? It won't press the tool buttons, it just keeps planning?.
0
0
1
Ingested some legal/confidential documents using @llama_index. When I do QA over them. @OpenAI did its best to ignore them completely. Apparently adding CONFIDENTIAL to any document will effectively blind the llm. cc, @jerryjliu0 @lmeyerov.
1
0
3
To all the flowers hanging in a romantic awning here in Berlin, where are your diplomats?.
0
0
2
Real culture is the antidote to legal fictions and other forms of government.
0
1
5
RT @EricRWeinstein: On now discussing the “@joerogan experience” discussion on the theories of @terrencehoward.
0
88
0
Watching my trading algo give a very confident "sell now" signal and @coinbase says "not today shmedrick".
2
0
4
Thanks @jupyterthon and @Cyb3rWard0g for hosting CEO and founder of @Graphistry, @lmeyerov talking about what we’ve been building with @LOUIE_AI and the future of GenAI “thinking” notebooks. As bonus, see how we won the #cybercomm AI challenge, and how
louie.ai
Agentic Automation for Force Multipliers. Enterprise-Grade Security. Bring your team and data into the era of agentic investigation & AI automation.
0
2
6
SymbolicAI makes LLM engineering easy, intuitive and logical. Check out Thanks @DinuMariusC.
huggingface.co
1
3
4
RT @DinuMariusC: 🚀 SymbolicAI – a framework for logic-based approaches combining generative models and solvers. Alongside, we introduce a b….
arxiv.org
We introduce SymbolicAI, a versatile and modular framework employing a logic-based approach to concept learning and flow management in generative processes. SymbolicAI enables the seamless...
0
61
0
Market research in the age of GPT APIs: OpenAI can swap out a model, making tests fail across the thousands of teams racing to product, and then measure the response via twitter. #domino #distributedcompute.
0
0
1
It's a bizarre thing to have dedicated years of my life learning how to code merely as a means to return to poetry. #Promptshare.
0
0
8
RT @Graphistry: Got logs? Our graph AI for event data hands-on at @NVIDIAGTC kicks off in 1 hour:. With CEO @lmeyerov & Head of AI @silksp….
0
1
0
RT @mijaspao: Thanks @lmeyerov and @silkspaceships for this amazing workshop. Looking fordward to put it all into practice at our @Graphis….
0
1
0
ChatGPT doesn’t have help desk built into the UI. I can’t ask it why I’m not seeing chat convos on the left hand side anymore. Ideally prompting “instructions: read latest release CHANGELOGS, website, marketing, etc and explain how changes there affected the UI” under the hood.
0
0
2
RT @lmeyerov: OH @Graphistry (*cough* ok @silkspaceships at product sync):. GPT today has a 4K token limit which feels like being in 1980….
0
3
0
Finding linear relations in Feynman integrals converts their calculations to a linear algebraic problem. Neat.
0
0
0