on Llm, Fine-tuning, Machine learning, Ai, Lora, Qlora, Openai, Hugging face
Prompt engineering gets you 80% of the way. Fine-tuning gets you the last 20%—but it’s expensive and easy to mess up. Here’s when it makes sense and how to do it right.
on Graphql, Federation, Apollo, Microservices, Api gateway, Supergraph, Typescript
REST APIs served us well, but they don’t compose. GraphQL Federation changes that—each team owns their subgraph, and the router stitches them into one unified API.
on Edge computing, 5g, Cdn, Cloudflare workers, Aws wavelength, Latency, Iot
Edge computing isn’t new, but 5G is making it practical. With sub-10ms latency becoming achievable, we can finally build real-time applications that were impossible before.
on Webassembly, Wasm, Wasmtime, Wasmedge, Spin, Fermyon, Serverless, Rust
WebAssembly started as a way to run C++ games in browsers. In 2026, it’s becoming a serious option for server-side workloads—fast startup, strong sandboxing, and true portability.
on Rust, Backend, Web development, Axum, Tokio, Api, Performance
Rust has been “the next big thing” for backend development for years. In 2026, the ecosystem has matured significantly. Is it finally ready for your next API?
메모는 기억보다 확실하다