Bạn có thể thích
little little goals: & spending 2 hours practising new things & spending 2 hours on kaggle everyday & publishing blogs twice a week & small contributions to open source & losing 2 kilos per month -> (24 kilos this year) & finding my optimum sleep cycle (through long experiments)
if this is you i would take the following very seriously: what worked for me was teaching. you get obsessed with a paper or architecture or whatever it is and have the pain point of not being able to find good resources on it. whether it be that for LLMs 3 years ago, or CUDA one…
@elliotarledge Any suggestions about projects, papers that I should study to give myself a good chance in ML field? I have intermediate level knowledge of about most of ML related domains.
Don’t overthink AI agents. > Learn Chain-of-Thought (CoT) > Learn Tree of Thoughts (ToT) > Learn ReAct Framework > Learn Self-Correction / Reflection > Learn Function Calling & Tool Use > Learn Planning Algorithms (LLM+P) > Learn Long-term Memory Architectures > Learn…
This mindset changed by life. Maybe it'll change yours too. Do it badly.
- build an autograd engine from scratch - write a mini-GPT from scratch - implement LoRA and fine-tune a model on real data - hate CUDA at least once - cry - keep going the roadmap - 5 phases - if you already know something? skip - if you're lost? rewatch - if you’re stuck? use…
i built a simple tool that makes Claude Code work with any local LLM full demo: > vLLM serving GLM-4.5 Air on 4x RTX 3090s > Claude Code generating code + docs via my proxy > 1 Python file + .env handles all requests > nvtop showing live GPU load > how it all works Buy a GPU
Retrieval techniques I’d learn if I wanted to build RAG systems: Bookmark this. 1.BM25 2.Dense Retrieval 3.ColBERT 4.DPR (Dense Passage Retrieval) 5.ANN Indexes (FAISS, HNSW) 6.Vector Quantization 7.Re-ranking (Cross-Encoder) 8.Late Interaction Models 9.Embedding Normalization…
Inference optimizations I’d study if I wanted sub-second LLM responses: Bookmark this. 1.KV-Caching 2.Speculative Decoding 3.FlashAttention 4.PagedAttention 5.Batch Inference 6.Early Exit Decoding 7.Parallel Decoding 8.Mixed Precision Inference 9.Quantized Kernels 10.Tensor…
You're in an ML Engineer interview at Databricks The interviewer asks "Your production chatbot's accuracy was 95% at launch. Six weeks later, user complaints are up and evals show 80%. What do you do?" You reply : "The model is wrong, we need to retrain it." Game over.…
Google's former CEO Eric Schmidt shares his weekend habit that led to billion dollar decision.
This has been the secret to my entire career
Very interesting share from my team: @ChrisDeotte @The0Viel @0verfit @Giba1 ) Lesson from competitive ML on @kaggle. Kaggling is not all you need to do to become great data scientists or ML engineers But it is extremely useful for improving your modeling skills. Hope that…
After hundreds of Kaggle competitions and years of trial, error, and wins, our team @ChrisDeotte @The0Viel @0verfit @giba1, have compiled a playbook for tabular modeling—a system that’s carried us to the top of leaderboards and held up on real-world data. developer.nvidia.com/blog/the-kaggl…
Some resources about GPUs I found good as a noob in GPU programming, 1. jax-ml.github.io/scaling-book/g… 2. modal.com/gpu-glossary/p… 3. multimodalai.substack.com/p/the-mlai-eng… 4. bytesofintelligence.substack.com/p/maximizing-g… 5. youtube.com/playlist?list=…
As software engineers, most of us have an itch to improve - to become better at coding, architecture, or leadership. But do you see the problem? It's all fuzzy and vague. I follow a simple flow of questions that helps me stay focused, improve consistently, and make an…
We spent the last year evaluating agents for HAL. My biggest learning: We live in the Windows 95 era of agent evaluation.
"HDM achieves competitive 1024x1024 generation quality while maintaining a remarkably low training cost of $535-620 using four RTX5090 GPUs"
Hugging Face 🤗 cooked again. They just dropped a new free course with certification on fine-tuning: > instruction tuning > RL and preference alignment > evaluation > creating synthetic data By the end of 2025 you have learned these essential skills from the best.
Pure systems programming talk on how deep we have to go to understand issues and solving it! Bliss to watch !
At @modal we've built every layer of the AI infra stack from scratch — from filesystems and networking to our own async queues and multi-cloud GPU orchestration. I sat down with @narayanarjun from @amplifypartners to go into depth on all of this, including the fun ways the…
You have all the time you need, you're just spending it poorly. Don't tell me you don't have time for Linux or kids OR BOTH. You have time for all of it once you stop filling your day with junk activities.
anyone else getting sick and disgusted of the state of social media / phone addiction and how we let it infiltrate every single corner of society? when i take my daughter on the playground there are toddlers clutching phones bigger than their heads.. most of the time they sit on…
United States Xu hướng
- 1. Brian Kelly 14.8K posts
- 2. #VERZUZ 54.1K posts
- 3. Yamamoto 65.3K posts
- 4. Birdman 14.5K posts
- 5. No Limit 34.9K posts
- 6. Colorado 33.9K posts
- 7. Mia X 574K posts
- 8. Cash Money 29.4K posts
- 9. Deion 1,954 posts
- 10. Wayne 37.3K posts
- 11. Salter 1,672 posts
- 12. Texas A&M 13.8K posts
- 13. Death Valley 5,333 posts
- 14. Michigan 38.7K posts
- 15. Coach Prime 1,327 posts
- 16. #GoBlue 3,638 posts
- 17. Aggies 5,201 posts
- 18. Turk 569K posts
- 19. #HalloweenHavoc 51.2K posts
- 20. Baton Rouge 3,058 posts
Bạn có thể thích
Something went wrong.
Something went wrong.