on Vector database, Ai, Rag, Embeddings, Pinecone, Weaviate, Machine learning
Every AI application needs a memory. When you ask an LLM about your documents, how does it find relevant information? The answer is vector databases—specialized systems that store and search high-dimensional embeddings at scale.
on Rust, Backend, Api, Performance, Web development, Axum, Tokio
Rust has crossed the chasm. What started as a systems programming language is now powering web backends at Discord, Cloudflare, AWS, and countless startups. In 2026, the ecosystem is mature enough that choosing Rust for a new backend is no longer a risky bet.
on Platform engineering, Devops, Developer experience, Idp, Backstage, Kubernetes
Platform engineering has emerged as the answer to DevOps complexity. Instead of expecting every developer to be a Kubernetes expert, platform teams build internal developer platforms (IDPs) that abstract away infrastructure while maintaining flexibility.
on Kubernetes, Opentelemetry, Observability, Devops, Monitoring, Distributed systems
Debugging distributed systems is hard. Requests bounce between dozens of services, and when something breaks, finding the root cause feels like searching for a needle in a haystack. OpenTelemetry (OTel) solves this by providing a unified standard for collecting traces, metrics, and logs.
on Ai, Agents, Llm, Claude, Gpt, Automation, Machine learning
AI agents have evolved from experimental toys to production-ready systems. In 2026, they handle customer support, code reviews, data analysis, and complex workflows autonomously. Here’s how to build agents that deliver real value.
메모는 기억보다 확실하다