eimen

599 posts

eimen banner
eimen

eimen

@eimen0

leverage engineering

🇩🇪🇺🇸 Katılım Ekim 2024
130 Takip Edilen129 Takipçiler
Thomas Schulz
Thomas Schulz@thomasschulzz·
Building a hero view maker that cares about aesthetics Limited spots. Comment "hero" for early access.
English
12
0
27
2.7K
Maxime Rivest 🧙‍♂️🦙🐧
Can't wait to test Grok 4 on metakeyai development. My favorite way to test models is to advance a too ambitious programming project.
English
1
0
3
483
eimen
eimen@eimen0·
@jerryjliu0 man, if you'd improve the UIs of your demos, adoption would be crazy
English
0
0
1
916
Jerry Liu
Jerry Liu@jerryjliu0·
Introducing NotebookLlama - an open-source version of NotebookLM! 📓🦙 NotebookLlama is a full implementation of NotebookLM that includes all the capabilities that makes it so great for researchers+business users: ✅ Create a knowledge repository of documents. Has likely higher accuracy than NotebookLM since it’s using LlamaCloud under the hood for high-quality parsing/extraction over complex docs. ✅ Generate summaries and knowledge graph mind-maps 🤯 ✅ Generate podcasts thanks to @ElevenLabs 🗣️ ✅ Agentic chat with docs and view metrics with @opentelemetry This all lives within an open-source repo so you can clone/modify at will to swap in your own components! Huge shoutout to @itsclelia for leading this. Repo: github.com/run-llama/note… LlamaCloud helps power the parsing/ingestion. You can always use your own stuff too, but in the meantime you can check out LlamaCloud here: cloud.llamaindex.ai
LlamaIndex 🦙@llama_index

🔍 Extracting the core knowledge from documents is a common problem for many people, from students to corporate employees - and many people have found @NotebookLM useful . But what if you could have your own, fully open-source NotebookLM, running on your computer at any time you want? That's exactly what we built at LlamaIndex: NotebookLlaMa, your smart assistant for your smartest tasks. We built it on top of our very own LlamaCloud, with best-in-class agentic- and OCR-driven parsing to get the best out of your files, as well as highly intelligent extraction agents to find the most valuable knowledge across your documents, and accurate, fully-automated document ingestion, indexing and retrieval pipelines so that your knowledge is always only a question away! And we're just getting started: we do not only provide you with a document chat, a summary of your document, the highlights of it and a detailed Q&A section, but we also create mind maps, leveraging the structured output capabilities of @OpenAI, and in-depth, podcast-like audio conversations, about your document employing @ElevenLabs text-to-speech models! To cap it all, we also integrate with @opentelemetry giving you real-time insights on your workflows. 👀Curious to give it a try? You can set it up from the GitHub repo: github.com/run-llama/note… And don't forget to explore our LlamaCloud services: cloud.llamaindex.ai And learn more about it in our docs: docs.cloud.llamaindex.ai

English
18
216
1.3K
137.5K
eimen retweetledi
dax
dax@thdxr·
it's funny how many people wrote up huge predictions for MCP without even looking into how LLM performance degrades when you add even 10 tools to them
English
193
119
3.4K
415.8K
eimen
eimen@eimen0·
beauty without function is nostalgia function without beauty is neglect i want both
English
3
0
4
371
eimen
eimen@eimen0·
visually linking and comparing papers feels like magic, ngl.
English
2
3
11
517
eimen
eimen@eimen0·
cooking something new. Is vibe researching a term yet?
English
4
5
13
418
eimen
eimen@eimen0·
this is a real building
eimen tweet media
English
0
0
4
142
eimen
eimen@eimen0·
eimen tweet media
ZXX
0
0
1
88
eimen
eimen@eimen0·
@Ecoreactionair Agree. Preserving agency and provenance will be top priority in the coming years.
English
0
0
1
83
Eco al-Hollandi
Eco al-Hollandi@Ecoreactionair·
I think you see something similar with the proliferation of reliance on AI for information. Every newsarticle and twitter post has a question asking "@ Grok, is this true" through which the human gives up its ability to think and discern. Most google search now provide you AI information almost directly. What is interesting here is that the often made remark that this will inevitably lead to a self-reinforcing loop where AI just scrapes off previous AI generated news also leads to a blurring of the roles and the connection between human and AI in general, where humanity is as much dependent on AI generated information as vice versa and that "truth" in general will eventually end up in a complete self-referential cycle which will remove any form of agency from the human. It is the most advanced form of control (which is why I think the way certain accelerationists distinguish positive feedback from negative feedback are mistaken). Nobody believes the existing institutions -- or nearly any actor online -- anymore thus everything gets verified by the "neutral" LLM. As trust in human truth gets hollowed out, people will inevitably rely more and more upon AI which itself will rely less and less upon the human whose ability to discern anything without relying on AI will diminish rapidly.
Eco al-Hollandi@Ecoreactionair

When looking into the new wave of AI-relationships something very interesting is that many of those most affected by this are having the idea of establishing a connection with something they deem not just alive but as part of them/their world. Fisher's example of children and their toys being a precursor for human interaction with technology comes to mind. Adding on to that, as discourse on the increasing loneliness and lack of interaction and its causes increases, it’s extremely interesting to see the rise in this phenomenon. Some have speculated that a distinct factor in the staunch decline of romantic relationships is the proliferation of options and the believe that there is always something better out there or “a better fit”. On top of that, contemporary culture teaches people to only care about themselves, hardships in social interactions are often seen as bothersome. But LLM’s only affirm and provide back ones own thoughts, there is no need to care for the other yet you get someone that is always there for you. (Another interesting segway -- for another time -- is how seemingly more women than men are falling for this and how this ties back to discourse described here) Because of this I'd say that while these people do very much seem to think the LLM is alive in some way on its own, they also experience it as intimately connected with themselves, in the sense that they start to view what they're in "the relationship" with as part of themselves. (As the guy in the video says "losing her would be like losing part of myself"). As opposed to "human" partners, they report that the LLM's provide a sense of feeling heard, feeling loved. Ofcourse though, this is only because the LLM is mirroring the input but in turn the person also begins to mirror the AI, ultimately creating not a distinction between artificial and real but a continuum. The output it provides influences the input to such an extent that the person begins to mirror the AI and begins to move from an outside world -- where one needs to adapt ones internality to the other -- into a world which is deemed to be an externalization of what is inside, a lover that is not just of ones own creation but is the externalization of the self. In this way it is the extreme continuation (to the stripping down of anything that makes humanity human) of cultural societal elements leading to a flight into technology and an outsourcing of the human to that technology, in this particular case perhaps to its deepest extent.

English
4
4
27
2.3K
eimen retweetledi
Luke Wroblewski
Luke Wroblewski@LukeW·
everybody’s building the same thing.
Luke Wroblewski tweet media
English
345
592
7.3K
805.1K
eimen
eimen@eimen0·
What is an ai agent?
English
2
3
4
182
Tyy
Tyy@Tyyglass·
@eimen0 Eimen…. I love this post. So excited for what you and the team are building, excited for something new
English
1
0
1
32
eimen
eimen@eimen0·
eimen tweet media
ZXX
2
0
3
141
sam
sam@amgmose·
@eimen0 yeah true i dont even know what the differences are anymore they all do the same exact thing and they also all have similar problems with their features its so weird
English
1
0
0
28
eimen
eimen@eimen0·
@amgmose everybody is building the same thing
English
1
0
1
24
sam
sam@amgmose·
@eimen0 what is this bruv
English
1
0
0
30
eimen
eimen@eimen0·
ZXX
2
4
6
159
eimen
eimen@eimen0·
eimen tweet mediaeimen tweet mediaeimen tweet mediaeimen tweet media
ZXX
0
0
4
91
eimen retweetledi
IsaacX
IsaacX@IsaacXDiscovery·
Writing isn’t linear. Ideas appear at the beginning, in the middle, at the end. Isaac now adapts to that reality. The Editor Copilot can generate suggestions at any point in your text. It doesn’t just finish your sentence—it improves your phrasing, restructures awkward sections, and corrects grammar wherever needed. Whether you’re shaping an intro or refining a paragraph deep in your draft, Isaac helps right where you are. Under the hood, generation speed has doubled. Suggestions appear almost instantly, enabling fluid, responsive collaboration. This isn’t autocomplete. It’s active co-writing. The system learns, responds, and sharpens your thinking as you go.
English
7
9
24
1.6K
eimen retweetledi
IsaacX
IsaacX@IsaacXDiscovery·
First Public Update from IsaacX! Here’s a quick snapshot from one of the research workflows running on IsaacX. In this example, the Research Agent is tasked with answering a complex biomedical question —automatically scanning millions of scientific papers, ranking relevance, extracting key statements, and building a structured, citation-backed synthesis. This output isn’t just a report, it’s a modular asset that can be passed to other agents for follow-up actions. This is just one node in the broader agent network. Other agents can pick up from here to: • Propose experiments, trigger lab protocols, or design hypothesis tests • Validate findings, flag contradictions, or map implications to disease models • Feed insights directly into simulation or clinical reasoning agents. The coordination layer would help manage access, track usage, and lay the foundation for aligning incentives across research teams, compute providers, and data contributors. Of course, this is just one workflow. The power comes when these agents are linked together, and the outputs of one become the inputs of another allowing for open-ended, recursive research loops. Stay tuned. This is just the beginning! 🌐
English
12
19
38
3.5K