Frontend Engineer
yeet
Frontend Engineer
yeet
Real-time kernel observability with eBPF
yeet is built by a tight-knit group of systems, observability, and kernel nerds who refuse to accept “mystery outages” as normal. We’re a small, high-impact team with domain expertise in Linux, eBPF, performance instrumentation, and developer tooling.
We move fast, ship often, and take deep ownership. If you enjoy operating close to the metal, building tools that real engineers lean on during 3 AM incident sprints, and solving hard problems with minimal overhead, you’ll feel right at home here.
We’re looking for a Frontend Engineer to push the limits of what’s possible in the browser.
yeet’s UI is a canvas-based app that ingests and visualizes huge streams of data. That means thinking beyond the usual component frameworks: Architecting solutions with data pipelines, rendering loops, and memory constraints in mind.
You’ll be closer to the browser’s low-level engine than most frontend roles, making decisions that directly impact how engineers experience yeet in the mid-incident.
You’ve built performance-critical frontends where smooth rendering and responsiveness weren’t optional.
You know when React helps and when to drop down to Canvas, WebGL, or raw DOM-manipulations for speed.
You have a strong grasp of data structures and algorithms.
You write expressive TypeScript and use the type system to catch bugs before they ship.
You’re fluent in low-level web APIs (e.g. OPFS, OffscreenCanvas, Web Workers) and nerd out over the MDN docs.
You’re motivated by measurable performance: frame budgets, memory profiles, flame-graphs, not just “it feels fast.”
Experience with WebAssembly (WASM) for accelerating data-heavy or compute-bound tasks in the browser.
Familiarity with bundling and build tooling (Webpack, Vite, Rollup) to optimize large-scale apps.
Background in usability / UX for data-dense applications — making complex interactions intuitive without sacrificing power.
Deep knowledge of React internals including how the scheduler works and when to bypass it.
Yeet offers a real-time system observability platform built on eBPF that lets engineers inspect Linux kernel behavior in the moment using customizable probes and policies, with JavaScript scripting. It collects kernel events via eBPF, streams live data to the browser, and stores it in a local DuckDB so you can run SQL queries; users manage probes and visualize outputs through a CLI, YPM, and a web interface, with probes executing in secure V8 isolates. Unlike traditional tools that rely on logs and metrics after the fact, Yeet provides live, kernel-level visibility and remote deployment capabilities. The goal is to give site reliability engineers and system administrators immediate answers to performance questions and enable live troubleshooting when dashboards aren’t enough.
Explore more opportunities at yeet