SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Stream is a Series B company serving over a billion users with billions of API requests monthly and a 99.999% uptime SLA. We're hiring a Staff Software Engineer for our Infrastructure team to build the platform that powers all Stream products.
This year, Stream is completing a major infrastructure migration: simplifying hosting and rebuilding the data layer on self-hosted, open-source systems that are faster and cheaper to operate. The next phase is building the infrastructure platform every Stream product runs on, with real SLOs, transparent costs, sustainable on-call rotations, and tools that help teams ship faster.
You will shape how much Stream spends and how fast it grows, with significant autonomy from your first weeks. This is a software engineering role first—attitude, ambition, and willingness to experiment across the stack matter more than specific tool expertise. AI will handle much of the DevOps-heavy work; what we need from you is system design judgment, strong fundamentals, and clean code.
The Infrastructure team is small and senior, moving fast and sometimes hectic. There are no large-org support structures here.
Key responsibilities:
- Define SLOs and success metrics for the platform. Set latency and uptime targets for every product team, and decide what we measure (error rates, delivery times, customer app health).
- Lead major infrastructure decisions. Design reviews and decisions on where and how Stream runs, working across backend, video, and moderation teams on cross-service tradeoffs.
- Build platform tooling. Write internal tools and services that help engineers deploy and debug faster, plus documentation and AI skills that make launching new products repeatable.
- Own on-call and incident response. Participate in rotations and ensure every root cause leads to durable fixes, not workarounds.
- Make cost a first-class engineering concern. Weigh cost alongside reliability in every decision and give teams the visibility to do the same.
What we're looking for:
- A software engineer who has built systems, not only configured them. You write production code in Go, Python, or similar languages. Scripting-only backgrounds are not a fit.
- Strong system design skills. You reason about failure modes, capacity, and cost in distributed systems and can explain your choices.
- Solid fundamentals in networking, storage, and concurrency. You ask why systems behave as they do rather than accepting defaults.
- Ownership of high-scale production systems and ability to set technical direction beyond your own work through design reviews, RFCs, and decisions others build on.
- Experimental mindset. When you encounter unfamiliar Kubernetes, databases, or cloud services, you dig in, use AI to close gaps, and ship.
- Daily use of AI tools in your engineering work.
- Ability to treat cost and reliability as engineering problems and quantify outcomes you've driven.
- Clear writing skills. Much of this role involves documentation and standards that other teams will follow.
- Comfort in small teams, setting direction and reviewing PRs in the same week.
Nice-to-have experience:
- Kubernetes in production: cluster architecture, workload design, or migrations you've led.
- PostgreSQL at scale, ideally self-hosted or with CloudNativePG: sharding, replication, partitioning tradeoffs.
- Valkey or Redis at scale, ideally on Kubernetes.
- Hands-on AWS and GCP experience, including multi-cloud setups or provider migrations.
- Cloud commitment and reservation strategy (committed use discounts, savings plans) or formal FinOps practice.
- SLO design, alert hygiene, and Prometheus/Grafana observability stacks.
- Real-time systems: WebSockets, WebRTC, streaming, or persistent-connection workloads.
- Open source contributions or writing/talks on cloud, platform, or distributed systems.
Tech stack: Go, gRPC, RocksDB, Python, PostgreSQL, RabbitMQ, GCP, Grafana, Prometheus, ELK, Jaeger, Tempo, Datadog, Redis, Memcached, Claude Code, Cursor.
You'll thrive here if you want platform and infrastructure problems at scale most engineers never touch, with autonomy to set direction. You prefer trying, measuring, and iterating over waiting for perfect plans. You ship and learn fast, even when hectic. You're self-directed and comfortable in globally distributed teams.
You probably won't fit if you want to specialize in one stack layer, prefer tightly scoped tickets with step-by-step direction, or need a calm, highly predictable environment.
