M
@_mihirpatel67
|| ॐ || Data Geek || #AI #ML 👨🏻💻 Data Scientist || I 😍 Reading
You might like
Come check out Agent Laboratory at #EMNLP2025! We will discuss how LLM Agents can help accelerate scientific discovery. I'll be presenting on Nov 5 @ 8:00. Looking forward to the discussions!
The Illustrated NeurIPS 2025: A Visual Map of the AI Frontier New blog post! NeurIPS 2025 papers are out—and it’s a lot to take in. This visualization lets you explore the entire research landscape interactively, with clusters, summaries, and @cohere LLM-generated explanations…
This thought is converging from many sides. Transformer based LLMs are not going take us to human level AI. That famous Yann LeCun interview. "We are not going to get to human level AI by just scaling up MLMs. This is just not going to happen. There's no way. Okay, absolutely…
Fei-Fei Li (@drfeifei) on limitations of LLMs. "There's no language out there in nature. You don't go out in nature and there's words written in the sky for you.. There is a 3D world that follows laws of physics." Language is purely generated signal.
Derivation of time-dependent Schrödinger's equation using wave mechanics ✍️
💥 Top 50 LLM Interview Questions & Answers Get ready for your next AI / ML / LLM interview with this power-packed Q&A guide covering: ✅ Prompt Engineering ✅ Fine-tuning & RAG ✅ Transformer Architecture ✅ Tokenization & Attention ✅ Real-world LLM Scenarios Perfect for…
Highly recommend this work from @shannonzshen if you're interested in human–agent collaboration but have felt discouraged by how hard it is to quantitatively study collaboration with humans in the loop. To quantify human-agent collaboration, the work introduces…
Today's AI agents are optimized to complete tasks in one shot. But real-world tasks are iterative, with evolving goals that need collaboration with users. We introduce collaborative effort scaling to evaluate how well agents work with people—not just complete tasks 🧵
If you want to increase your surface area for luck, focus on producing proof of work. You will make yourself a bigger target for luck.
Stop Prompting LLMs. Start Programming LLMs. Introducing DSPy by Stanford NLP. This is why you need to learn it:
step-by-step LLM Engineering Projects each project = one concept learned the hard (i.e. real) way Tokenization & Embeddings > build byte-pair encoder + train your own subword vocab > write a “token visualizer” to map words/chunks to IDs > one-hot vs learned-embedding: plot…
This broke my brain. A team at Sea AI Lab just discovered that most of the chaos in reinforcement learning training collapse, unstable gradients, inference drift wasn’t caused by the algorithms at all. It was caused by numerical precision. The default BF16 format, used across…
This guy literally teaches you how to learn complex skills
Simplicity is at the heart of great software. This is one of the reasons why Claude Code has been sticky for me. As a builder, I love planning and brainstorming, and this is now a key focus of Claude Code. I use Shift + Tab a lot to cycle between brainstorming, planning, and…
Elon’s famous analogy: “Every person in your company is a vector. Your progress is determined by the sum of all vectors”. At big companies, vectors often point to different directions. The sum is zero. xAI has a small team of engineers, but every vector points at the same…
bf16 doesn’t work when you do sft on qwen-3-1.7b on a multi-turn wordle env. reward stays at 0.2 whereas with fp32 it goes to ~0.65.
ai engineer interview question
Fei-Fei Li says language models are extremely limited. This @GoogleDeepMind paper makes almost the same point, just in the world of video. The models are just very advanced pattern matchers. They can recreate what looks like reality because they’ve seen so much data, but they…
Fei-Fei Li (@drfeifei) on limitations of LLMs. "There's no language out there in nature. You don't go out in nature and there's words written in the sky for you.. There is a 3D world that follows laws of physics." Language is purely generated signal.
i used the following resources to understand the quantization process for NVFP4, the paper below made it very clear. - cublas docs: docs.nvidia.com/cuda/cublas/in… - arxiv paper (appendix B): arxiv.org/pdf/2509.25149 - OCP microscaling spec: opencompute.org/documents/ocp-…
LoRA by Hand ✍️ ~ How does LoRA make fine-turning efficient (fewer parameters)? I drew this animation to show you. Pay attention to the number of red cells. On the left, regular fine-tune would involve 10x10=100 trainable parameters to update the original weight matrix.. On…
United States Trends
- 1. $APDN $0.20 Applied DNA N/A
- 2. $SENS $0.70 Senseonics CGM N/A
- 3. $LMT $450.50 Lockheed F-35 N/A
- 4. Good Friday 36.7K posts
- 5. #CARTMANCOIN 1,979 posts
- 6. yeonjun 276K posts
- 7. Broncos 68.4K posts
- 8. Raiders 67.1K posts
- 9. Blockchain 200K posts
- 10. #FridayVibes 2,608 posts
- 11. #iQIYIiJOYTH2026 1.48M posts
- 12. Bo Nix 18.8K posts
- 13. Geno 19.6K posts
- 14. Tammy Faye 1,817 posts
- 15. Kehlani 12K posts
- 16. daniela 57.8K posts
- 17. MIND-BLOWING 22.8K posts
- 18. #Pluribus 3,204 posts
- 19. John Wayne 1,151 posts
- 20. Danny Brown 3,379 posts
Something went wrong.
Something went wrong.