Deploying AI Models on Kubernetes
A practical field guide to GPU node pools, model serving with vLLM and Triton, and the dark art of autoscaling inference workloads on GKE and EKS — without setting your cloud bill on fire. (Day 22)
Kubernetes96
Check out the latest posts
Open-Source Models: Llama, Mistral & Running Locally and on the Cloud
Self-hosting LLMs Local Machines and on EC2/GKE with Ollama — when open-source beats API services (Day 15)
Embeddings & Vector Databases Demystified
What vectors are, why semantic search matters, and how Pinecone/Weaviate/pgvector fit in (Day 8)
AI Safety, Hallucinations & Trust in Production
Why AI lies confidently and how to build guardrails as an infrastructure problem (Day 7)
Tokens, Context Windows & Why They Matter for Cost
How to think about context limits, pricing models, and request optimization (Day 6)
AI APIs: OpenAI, Anthropic, Google & AWS Bedrock
Comparing model APIs like you compare managed services (Day 5)