Software Engineer, Streaming SQL
Polygon.io·Mid-level·Remote · United States
Posted Oct 27, 2026·polygon.io.com
About The Role
At Massive, our live data products keep users current on financial markets, processing terabytes of data a day while serving results that are fresh, predictable, and fast.
fireflow is the streaming dataflow engine behind those products. You write SQL that keeps running as new data arrives, and fireflow keeps the results correct and up to date. It’s built on Apache Arrow and DataFusion, with speed, correctness, and portability as first-class concerns. Deployments range from long-lived, stateful streaming jobs to WebSocket servers hosting thousands of stateless dataflows. Underneath are a disaggregated storage layer, a formally specified control plane, and an incremental changelog executor.
fireflow is still growing, and our small, highly technical team owns every layer, from the SQL parser to query execution. That takes engineers who think about databases from first principles and aren’t satisfied with “it seems to work.”
If you’ve ever watched a query recompute everything for each new event and thought there had to be a better way, you’ll fit in.
Responsibilities
- Design and build features across fireflow in Rust, owning them from first design through production
- Profile, benchmark, and optimize for throughput and latency
- Write design docs and RFCs, and review code from the rest of the team
- Test for correctness with deterministic simulation and formal modeling
- Stay up to date with stream processing and database research, and bring the best ideas into fireflow
Skills & Qualifications
We want engineers who think carefully about performance and reliability, own their systems end to end, and understand how databases work.
- Strong proficiency in Rust; safe, performant, idiomatic code comes naturally
- Solid systems programming fundamentals: you can reason about memory, concurrency, and performance from first principles
- Strong grounding in data structures and algorithms
- Backend experience on or alongside database systems, and curiosity about how they work; more depth is better
- The ability to explain why a change is correct, not just that it passes tests
- Bachelor’s degree in Computer Science or a related field, or equivalent experience
Nice to Have
- Experience with stream processing or real-time analytical systems (Flink, Kafka Streams, Materialize, RisingWave, or similar)
- Experience with deterministic simulation, fault injection, or formal methods
- Familiarity with Arrow, DataFusion, or columnar execution
- Background in financial data
How We Work
We encourage agentic software development and know AI writes much of the code now. You are still accountable for what you ship, and you should understand your changes well enough to defend them. “The AI said so” is not a rationale. Pull request conversations are between people, not pasted model output.
