Y Hacker News new | ask | show | jobs

user: covi created: 2012-09-05 karma: 201 https://github.com/skypilot-org/skypilot/ submissions: SkyPilot Sandboxes: Run Agent Code on Your Own Kubernetes, at Scale 6 points | 0 comments RL Doesn't Work on Slurm 5 points | 0 comments 0 points | 0 comments 0 points | 0 comments SkyPilot: One system to use and manage all AI compute (K8s, 20 clouds, Slurm) 15 points | 2 comments The Evolution of AI Job Orchestration 2 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments Large-Scale AI Batch Inference: 9x Faster Embedding Generation 1 points | 0 comments Abusing SQLite to Handle Concurrency 2 points | 0 comments 0 points | 0 comments Using DeepSeek R1 for RAG: Do's and Don'ts 13 points | 0 comments Building Large-Scale Image Search Using VectorDB and OpenAI Clip 3 points | 2 comments Fungible cloud compute and storage: AI training on any cloud 2 points | 0 comments Getting $1M cloud credits for AI startups – and using them wisely 1 points | 0 comments You Ran the Operational Database on What? Testing Spot Instances 3 points | 1 comments Can Multi-Modal LLMs "See" Images? A Deep Dive with ASCII Art 3 points | 0 comments Embarrassingly parallel batch jobs on multiple clouds or regions 1 points | 0 comments 0 points | 0 comments Llama 3.1 finetuning on your own data and infra, without vendor lock-in 2 points | 0 comments Guide: Finetune Llama 3.1 on your infra 5 points | 0 comments AI on Kubernetes Without the Pain 2 points | 0 comments Dagster and SkyPilot: Orchestrating LLM training cost-effectively 3 points | 0 comments Inflection CEO Left, Became CEO of Microsoft AI 5 points | 0 comments SkyServe: A cost-efficient, multi-region/cloud library for serving GenAI models 4 points | 0 comments SGLang: Fast and Expressive LLM Inference with RadixAttention for 5x Throughput 2 points | 0 comments SkyServe: 50% Cheaper AI Serving on Any Cloud with High Availability 5 points | 0 comments