Program

Node's native type stripping is now the default path for .ts files, and the compiler itself is moving to a Go port for a step-change in speed. This talk is a practical migration guide: which syntax is erasable and which isn't (enums and namespaces still need work), how to drop ts-node and tsc dist/ pipelines, what --erasableSyntaxOnly buys you in CI, and where this still doesn't reach, browsers still need a build step, and this talk is explicit about that boundary.

A live Webpack-to-Vite migration on a real production monorepo, with before/after numbers for cold start, HMR and CI time. Covers where Rolldown and Oxc fit into the Vite ecosystem, why Biome's linting is fast enough to change how often you run it, and the honest list of plugins and loaders that don't have a clean equivalent yet.

A decision framework, not a popularity contest. What changes when Bun is backed by a different company than the one you deploy with; where Node's ecosystem compatibility still wins outright; what Deno's permission model buys you that the other two don't. Built around a case study of running all three behind one API gateway and what broke.

Agents are fast and consistent; including about making the same mistakes repeatedly. This talk builds a concrete review process rather than relying on vibes: type-level constraints that catch a whole class of errors before a human looks, custom lint rules tuned to your framework's conventions, property-based tests for agent-authored code, and where in the pipeline review actually becomes the bottleneck.

What it actually takes to let an agent operate safely across a large JS/TS monorepo: writing an AGENTS.md that an agent will follow, sandboxing and permission boundaries, running evals on agent-produced diffs in CI, and the failure modes specific to large codebases — stale context, framework-convention drift, and agents confidently violating architectural boundaries they were never shown.

Cloudinary Media is the part of the full-stack AI story most talks skip. A look at how agents are now generating, editing and routing images and video inside a production pipeline: structured transformation requests, agent-driven asset variants at scale, and the governance problem nobody's solved yet, what happens when an agent decides what your brand's imagery looks like.

Next.js team From the team building it: Server Components, Cache Components, Partial Prerendering and server functions, explained as the mental model the Next.js team actually uses internally, not the marketing version. What changed since the App Router shipped, what's still being figured out, and how to reason about a route's caching behavior without guessing.

Every major framework has converged on some version of fine-grained reactivity. React's compiler replacing manual memoization, Vue's Vapor Mode, Svelte's runes, Solid's signals from day one. A side-by-side comparison of how each framework actually implements it under the hood, what it buys you in render performance, and which mental model to carry with you when you move between frameworks on the job.

Next.js team Most teams aren't migrated yet, and most migration guides assume a greenfield app. A practical, DX-focused walkthrough of moving a real production app: what to migrate first, where data-fetching patterns break, how to keep a team shipping features mid-migration, and the honest list of things that are still awkward in the App Router today.

A postmortem-driven talk built around 2025's wave of npm compromises and framework-level RCEs. Covers provenance and trusted publishing, minimum-release-age policies for dependencies, Node's permission model in practice, and a concrete checklist for auditing what's already running in your supply chain today.

AI features break observability's usual assumptions: a single user action can now fan out into multiple model calls with unpredictable latency and cost. This talk covers end-to-end OpenTelemetry tracing across browser, server and LLM calls, setting cost and latency budgets per feature, and catching silent failures (a tool call that returns the wrong shape, a model that "succeeds" with a bad answer).

With agents increasingly producing first drafts, what does seniority mean now judgment calls, architectural taste, review depth, mentoring and how should engineers be deliberately building those skills rather than letting them atrophy. Grounded in interviews with engineering leads making hiring and leveling decisions right now, not speculation.