Rust Performance in 2026 — Benchmarking, Profiling, and Real Wins
Practical Rust perf: criterion benchmarks, perf flamegraphs, dhat for allocations, common bottlenecks, and the discipline of measure-then-optimize.
Practical Rust perf: criterion benchmarks, perf flamegraphs, dhat for allocations, common bottlenecks, and the discipline of measure-then-optimize.
Practical Rust async traits: AFIT (async fn in traits), trait_variant for Send bounds, dyn-compatibility, and where async-trait crate still earns its keep.
Practical SQLx: query! macro vs runtime queries, migrations with sqlx-cli, transactions, FromRow, and the discipline that makes SQLx shine.
Production Rust CLI patterns: clap derive, async with tokio, progress bars with indicatif, TUIs with ratatui, config layering, and shipping cross-platform binaries.
Practical Axum middleware: auth, request ID, tracing, rate limiting, error mapping, and how Tower’s layer composition actually works.
Production Rust error handling. anyhow for binaries, thiserror for libraries, error wrapping, panic policy, and the patterns from real codebases.
Productivity Rust tooling for 2026. cargo, clippy, fmt, workspaces, nextest, watch, and the strict-lints that catch bugs in CI.
What async Rust feels like once it clicks — Tokio’s runtime, tasks vs futures, JoinSet, channels, cancellation, select!, and the shape of every concurrency pattern you’ll need on the backend.
A from-scratch production Rust HTTP service. Axum 0.8, sqlx with compile-time-checked queries, structured errors, request tracing, layered middleware, and the exact project layout I use.