Your curated collection of saved posts and media
Build custom retrievers that beat generic vector search by understanding domain-specific context and jargon 🎮 The team at @superlinked shows how to create a Steam games retriever using our custom LlamaIndex integration, combining semantic search with gaming-specific knowledge. 🎯 Custom retrievers hardwire domain expertise - perfect for gaming jargon and complex queries like "strategy co-op game with sci-fi elements" ⚡ Multi-field indexing combines name, description, genre into single searchable text for richer understanding 🔧 LlamaIndex BaseRetriever needs just one _retrieve method implementation to plug in any custom backend 🚀 @superlinked's InMemoryExecutor delivers sub-millisecond latency for real-time recommendations The guide walks through schema design, vector space configuration, and result processing with practical code examples. We show how combining @superlinked's mixture-of-encoders approach with LlamaIndex's RAG infrastructure creates retrievers that actually understand what users mean, not just what they type. Full tutorial with Colab notebook: https://t.co/hrEbW7wrib
https://t.co/hrWlaIPCg5
@fishPointer try @florafaunaai - example of what i'm working on rn https://t.co/jgG5W3bg6Y
@fishPointer try @florafaunaai - example of what i'm working on rn https://t.co/jgG5W3bg6Y
how do we keep track of rag updates, check out https://t.co/dACbePbAYo https://t.co/jlxcG28O9O
Claim: gpt-5-pro can prove new interesting mathematics. Proof: I took a convex optimization paper with a clean open problem in it and asked gpt-5-pro to work on it. It proved a better bound than what is in the paper, and I checked the proof it's correct. Details below. https://t.co/eNEGqyZG0L
making the youtube comment section happy has been the single hardest part of growing @latentspacepod https://t.co/4Y7eoh78HA
🆕 "RAG is Dead, Context Engineering is King" https://t.co/CUxBNvAuqi We asked @jeffreyhuber about the big @trychroma Cloud launch today, how Context Engineering is evolving, and why you should Look At Your Data (h/t @HamelHusain @jxnlco) full link to youtube pod below https://
making the youtube comment section happy has been the single hardest part of growing @latentspacepod https://t.co/4Y7eoh78HA
it's happening @jxnlco https://t.co/91rXUPqWuc
it's happening @jxnlco https://t.co/91rXUPqWuc
No amount of wealth will prepare you for discovering you've been accidentally paying for a Zoom subscription since 2021 https://t.co/sD3Ftv85UL
Looking for your next AI Engineering Role? Aa bunch of VC-backed startups are hiring founding engineers, in ai and backend systems. Don't miss out on some great opportunities and your next big challenge Apply now! https://t.co/1wqJdsRLpS
What does being on Comet feel like https://t.co/GhBTvqG5Or
@snibbmaster 20% compounding gains is exponential improvement. https://t.co/pvJuGN0uQK
This is true of any major benchmark (ARC-AGI, humanity’s last exam, time for task completion). If you compare to last month, things seem relatively stable. If you compare to last year, things seem to moving extremely fast. https://t.co/uonUxgEUMJ
Agents as UX bridges enabling mainstream users to access on-chain DeFi via integrated DEX/wallet experiences. @marouen19 https://t.co/tevrIk5fGt
Quantum computing now has an official AI assistant. Pleased to announce “Ask AI” for Q-CTRL, the $100mm+ company bringing quantum computing to reality. Their kapa announcement said it better than I could: “Abstraction is one of the biggest challenges in quantum computing”. Q-CTRL worked hard to build software abstraction layers on top of their quantum computing solutions. These are incredibly rich, deep concepts; Q-CTRL needed to make sure AI explained their docs correctly, and cited sources every step of the way. Now, developers for the next generation of compute can chat with an AI in natural language, and get back answers. Yes, even code generation. And yes, in many languages; Japanese, Korean, French, Spanish, and more. @qctrlHQ will onboard the next generation of Quantum Computing developers, and @kapa_ai is happy to assist. Try it out, and thanks to Q-CTRL for being an excellent customer!
Chain-of-Agents End-to-End Agent Foundation Models via Multi-Agent Distillation and Agentic RL https://t.co/A2ZwhHJYMa
discuss with author: https://t.co/nznP5GFAkf
We built the MOST consistent AI characters. Gave you FULL control in your storytelling. Higgsfield Soul is State-Of-The-Art. Always realistic, never plastic. And Soul ID KNOWS you better than anything. Retweet = 100 banger prompts in DMs. Character consistency for EVERYONE→ https://t.co/U3A0bB5BGe
MeshCoder LLM-Powered Structured Mesh Code Generation from Point Clouds https://t.co/dnfCdjcNWP
discuss with author: https://t.co/9yJ2adpInU
RynnEC Bringing MLLMs into Embodied World https://t.co/Du54RQXwuU
discuss with author: https://t.co/hy4QgRNhpb
Tinker Diffusion's Gift to 3D--Multi-View Consistent Editing From Sparse Inputs without Per-Scene Optimization https://t.co/APHW0CcUPj
discuss with author: https://t.co/BzFnKyKcOo
Oh yes! This is exactly what I wanted Audacity ported to web So cool: https://t.co/F5BfPTc2w9 https://t.co/p7SG1CIStb
@levelsio Audiomass my boy or Wavacity which is literally Audacity ported
Oh yes! This is exactly what I wanted Audacity ported to web So cool: https://t.co/F5BfPTc2w9 https://t.co/p7SG1CIStb
its test set is already available on @huggingface! definitely rare to find a music recommendation dataset, that is multimodal and conversational. so rare that there might be only one.. https://t.co/cXhCuIW0m0 https://t.co/BhI4vzH6C7
Tired: building AI for waifus & chatbots Wired: building AI for space exploration! Excited to introduce Surya, the first open-source AI foundation model for heliophysics, released by @NASA & @IBM on @huggingface! It's a 366M-parameter transformer model pretrained on 9 years (≈218 TB) of multi-instrument data from Nasa's Solar Dynamics Observatory, including 8 Atmospheric Imaging Assembly channels and 5 Helioseismic and Magnetic Imager products. It can be used to understand solar dynamics and predicting space weather which is critical for protecting satellites, power grids, communication systems, and astronauts. Model: https://t.co/pXxPb5UBOt Organization: https://t.co/UCB2W89zvz 🚀🚀🚀
In collaboration with @NASA, we've open-sourced Surya on @huggingface — a new foundation model designed to help researchers protect infrastructure through accessible, accurate modeling of space weather. It's going to totally change how we forecast solar storms. See how.🧵 https:/
Excited to release: Jupyter Agent 2 The agent can load data, execute code, plot results inside Jupyter faster than you can scroll! 🤖 Powered by Qwen3-Coder ⚡️ Running on Cerebras ⚙️ Executed in E2B ↕️ Upload your files All videos are in *real time*! https://t.co/AYnJJenGXN https://t.co/374nWL35LE
ByteDance just released the Seed-OSS 36B LLM on Hugging Face. It's an open-source model with powerful long-context, reasoning, and agentic capabilities. https://t.co/WP10Kuew6v
