Staff Software Engineer

Our mission is to reshape design by streamlining the path from concept to creation, bringing more impactful ideas into the physical world.

We’re well-funded (Series B $51m to date), growing double-digits MoM, and expanding the core engineering team in SF. The surface area is big: realtime collaboration, GPU inference at scale, a modern TypeScript stack, and serving real enterprise

Take end to end ownership of critical product areas, from architecture through launch, and set the technical direction for how we build at scale. You will drive design reviews, ship prototypes that push our product forward, and establish pragmatic engineering standards that balance speed and quality. Beyond your own contributions, you will mentor teammates into future tech leads and help us grow a world class engineering culture as Vizcom scales.

• Set the architecture roadmap to keep Vizcom reliable, scalable, and a joy to develop in as usage, models, and data grow.

• Ship hands-on: jump into code for critical paths, rapid prototypes, and hairy debugging (latency, memory, race conditions, WebGL/WASM quirks).

• Install quality guardrails that scale: test strategy, code-review norms, perf budgets, observability baselines.

• Mentor and multiply: develop senior engineers into tech leads, model clear written design, and practice crisp decision-making.

• Partner cross-functionally: represent engineering trade-offs with Product, Design, and Go-To-Market; act as the CTO’s delegate when needed.

• Reliability and performance: land an observability-driven plan for our top-5 critical flows; reduce p95 for at least one user-visible flow by about 30% through targeted fixes (query plans, caching, batching, streaming, GPU/CPU boundaries).

• Architecture: author one substantial RFC (for example, realtime doc model, rendering pipeline boundaries, or background job topology) and lead it to adoption.

• Team health: mentor two engineers with measurable growth in scope, design quality, or incident handling.

• Developer experience: remove one persistent friction point (local env speed, CI flakiness, or flaky e2e surface) and document the standard.

• High-fanout realtime collaboration (WebSockets and CRDT-style data models).

• GPU-backed image generation pipelines and queueing with bursty workloads.

• Performance across the stack: React and Vite bundles, GraphQL n+1s, Postgres query plans, caching at the right layer.

• Enterprise-grade SSO, SAML, SCIM, auditability, and secure multi-tenant boundaries.

• A design-forward UX that must still feel instant under load.

• Frontend: TypeScript, React, Vite, WebGL; realtime collaboration.

• API/Backend: TypeScript/Node, GraphQL (PostGraphile), Postgres, Redis, background workers.

• Infra: Kubernetes, Pulumi, CI/CD with GitHub Actions, Datadog for observability, feature flags.

• Security/Enterprise: SSO/SAML (WorkOS), SOC 2-minded practices.

Experience with every item is not required. What matters is that you are excited to shape it.

• You’ve led systems at scale in TS/Node (or similar) and know where complexity should live and where it should not.

• You are equally comfortable writing code and writing RFCs, and your reviews make everyone better.

• You communicate precisely in writing first, make trade-offs explicit, and keep momentum without hand-waving.

• You lead through influence, not title. People invite you to reviews because you raise the conversation.

• 8–12+ years building distributed web systems (TypeScript/Node preferred) with 2+ years at Staff scope.

• Depth with Postgres and performance/observability (profiling, traces, query tuning).

• Experience with container orchestration and production ops (Kubernetes or equivalent).

• Bonus: realtime collaboration models, WebGL/WASM, GPU workloads, or enterprise identity (SSO/SAML/SCIM).

• Small team, high ownership. Fewer people, more leverage.

• Design-centric. We obsess over speed and feel; “fast enough” usually is not.

• Write it down. RFCs, crisp decisions, living docs.

• Operate what you build. Everyone takes on-call support; incidents drive improvement, not blame.

• Intro (30m): mutual fit, role and impact.

• Architectural deep-dive (60–75m): a real Vizcom problem where you drive the trade-offs.

• Hands-on working session (90m): pair on a small, realistic slice.

• Cross-functional chat (30m): product and design collaboration.

• Founder/CTO conversation (30–45m): values, scope, expectations.

• References and offer.

If you want your work to be used daily by world-class designers and you like owning problems end to end, let’s talk. Apply with links (GitHub, LinkedIn, notable projects) and a short note on something hard you shipped.