Software Engineer

We're hiring Software Engineers (Staff → Principal) in Brooklyn and SF.Strobe is building the world's largest power plant — not a single site, but a distributed, fleet of buildings, batteries, EVs, and generators that buy and sell power in real time across wholesale energy markets.Power is now the binding constraint on AI and American reindustrialization. PJM, the largest grid, projects peak load rising ~21% to roughly 184 GW by 2030, driven largely by data centers. Reliability and cost increasingly depend on fast, controllable capacity sited near where power is consumed. That's what we build.What we've done in ~18 months: contracted capacity across 500+ households plus schools, gas stations, cold-storage facilities, office parks, manufacturers, car dealerships, hotels, and resorts. We connect those assets with our own hardware for telemetry and control, and run the fleet from a cloud control plane that forecasts load, solar, and price, estimates system state, and dispatches devices and bids into ERCOT, NYISO, PJM, and CAISO. Our services speak the Model Context Protocol (MCP), so LLM agents can act safely on live plant data.The team: some of the sharpest people I know from MIT, deep power-markets and finance experience from Goldman and hedge funds, operating chops from Samsara, and my cofounder — 15 years in power, most recently helping power the xAI Colossus data center.What you'd build. We're hiring across three levels, all hands-on:Staff, Platform — the cloud control plane and data layer: wholesale-market integrations, high-rate telemetry ingest, Green Button Connect pipelines, and event-driven AWS (Lambda, AppSync, EventBridge, IoT Core, TimeStream).Staff, Controls — the real-time edge: Rust on embedded-Linux RTUs, industrial protocol stacks (DNP3, Modbus, BACnet, MQTT), and failsafe dispatch logic where a bad command can trip a breaker.Principal — reliability and performance of the whole system end-to-end, and leading the integration of new capacity as the fleet scales by orders of magnitude.You'll fit if you: have shipped production code in Rust/Go/C++ and are fluent in Python or TypeScript; think from first principles and break big problems into small, shippable pieces; hold a high bar for reliability, operability, and security; and are comfortable on-call for systems that control physical hardware and move real money. Versatility beats narrow depth — we want people who plug in across controls, cloud, and UI.The work environment: small, high-performing team, massive individual agency, and AI at the core — an agent-native monorepo with deep investment in agent-enabled engineering.Comp: $180K–$280K depending on level, plus meaningful early-stage equity. Medical/dental/vision, flexible PTO, and access to prototyping shops (3D printers, CNC, electronics labs). On-site in SF or Brooklyn.