Senior Software Engineer 2 - Streaming and Durability

About the Role

Every transaction, every listing view, every agent interaction at Compass touches a streaming pipeline, a database, or a cache owned by the Streaming & Durability team. This team is responsible for the foundational data layer: the systems that have to be fast, correct, and isolated across thousands of tenants, every single time.

We don't just keep the lights on. We're solving hard problems: how do you run Kafka as a true platform, with clear authentication, managed schemas, and standard tooling, when it has run for years without a dedicated owner? How do you make Kafka upgrades, schema changes, and connector rollouts routine instead of risky? How do you enforce tenant-level data isolation across shared PostgreSQL clusters and Redis/Valkey caches without sacrificing latency? How do you build SDKs that make the safe path the easy path for hundreds of engineers who never think about infrastructure?

As a Senior Engineer, you'll own answers to those questions. You'll design systems that quietly protect millions of records, ship tooling that engineers across the company reach for by default, and operate infrastructure where the cost of getting it wrong is measured in trust, not just latency.

Responsibilities

- Own Kafka at Compass end to end, across our MSK and Confluent clusters: platform health, upgrades, capacity, and cost. Set the technical direction for how the rest of engineering uses streaming.

- Run the Kafka operational program: client authentication and access standards, Schema Registry ownership with PR-driven schema registration, Kafka Connect upgrades and standard management, producer compression defaults, and the runbooks and drills that let application teams recover quickly.

- Build and scale Kafka/MSK streaming infrastructure, the connective tissue between Compass services, including topic architecture, consumer patterns, partition rebalancing, upgrades, and dead-letter-queue and redrive guidance.

- Build multi-tenant data isolation into Compass's data layer, shipping SDKs in Go, Java, and Python that enforce tenant-safe access patterns so product engineers can move fast without risking cross-tenant data exposure.

- Support the database and caching fleet (RDS/Aurora PostgreSQL, ElastiCache Redis/ValKey, DynamoDB) alongside our database platform partners, including replication, failover, and capacity decisions.

- Drive production database upgrades at fleet scale, moving toward automation, with Terraform state changes across hundreds of clusters and zero customer-facing impact.

- Develop deep observability into the data layer and streaming platform, including SLI dashboards, tenant-dimensioned telemetry, and real-time performance insights that let the team (and the company) understand system behavior before customers do.

- Partner directly with product and platform teams to assess database readiness for scaling events, migrate workloads onto golden-path infrastructure, and resolve the hardest production incidents in the stack.

Basic Qualifications

- You've operated Kafka/MSK at scale and dealt with the hard parts: upgrades, consumer lag, partition rebalancing, and authentication (SASL/SCRAM or IAM).

- You know the surrounding Kafka ecosystem (Schema Registry, Kafka Connect) well enough to run it and to set standards for other teams.

- 7+ years of experience building and operating distributed systems at production scale in AWS.

- Working knowledge of a major managed database (PostgreSQL or similar), enough to partner with a dedicated database team and reason about the infrastructure around it.

- You've built shared tooling (SDKs, client libraries, platform APIs), and you understand what it takes to make infrastructure accessible and safe for engineers who aren't infrastructure specialists. Kafka client libraries are a plus.

- You think in trade-offs, not absolutes. You can articulate why you'd choose a soft application-enforced boundary over a hard infrastructure lock and when that calculus changes.

- You write Go (or can ramp quickly), with Java a plus, and you're comfortable in Terraform, Kubernetes, and IAM-heavy AWS environments.

- You've led cross-team operational programs, including change management, migration coordination, and rollback planning, where the blast radius of a mistake extends well beyond your own team.

- You can drive a technical decision end to end: frame the options, prove them out with benchmarks or a proof of concept, and lay out cost and risk for senior leaders.

Preferred Qualifications

- You've worked on multi-tenant isolation problems such as key prefixing, row-level security, and RBAC/ACL models and have opinions about where application-layer enforcement breaks down.

- You've built or contributed to observability platforms, specifically for database and cache monitoring.

- You've worked on Kafka cost and efficiency, such as compression, retention, and cross-AZ data transfer.

Compensation: The base pay range for this position is $176,000-$196,100; however, base pay offered may vary depending on job-related knowledge, skills, and experience. Bonuses and restricted stock units may be provided as part of the compensation package, in addition to a full range of benefits. Base pay is based on market location. Minimum wage for the position will always be met.