Hybrid 3 days (offices in NYC, Denver, CO and Charlotte, NC area)

Position Summary:

Judi Health is building the technology infrastructure our nation needs to deliver the healthcare we deserve — a domain where correctness is non-negotiable and latency has real-world consequences. The Architecture organization is responsible for the technical foundation the rest of the company builds on: the standards, patterns, and infrastructure primitives that engineering teams depend on to move fast and build correctly. The Realtime & Streaming team sits within Architecture's Core Platform function, owning the streaming infrastructure that services and data consumers across engineering depend on for timely, correct event delivery.

As a Staff Engineer in Architecture, your influence extends beyond what you ship directly. The standards you set, the patterns you establish, and the decisions you document become the reference point for how event-driven systems are built across the organization. You are building for the platform and for every team that builds on it.

The platform's consumers — cache invalidation systems, analytics pipelines, and change-data-capture subscribers — each carry different latency and ordering requirements. You will be directly response for scaling delivery without degrading it — honoring per-consumer SLAs as volume grows while keeping historical and real-time data in sync.

One of your first major initiatives will be leading the migration of our data warehouse synchronization layer from independently managed, batch-driven processes to a unified WAL-based realtime pipeline, with an open table format as the canonical source. This spans the full streaming path from WAL replication to the analytical systems downstream — Redshift and Snowflake — whose native streaming and ingestion capabilities

Position Responsibilities:

• Design, build, and operate the WAL replication and event streaming platform that cache invalidation, data warehouse synchronization, and change-data-capture consumers across the platform depend on for correctness and low latency
• Own the full lifecycle of the streaming platform — from replication topology and delivery guarantee design through operational recovery and post-incident review
• Define and own the performance and correctness contracts between the streaming platform and the services that depend on it
• Lead the migration of batch-driven data warehouse synchronization to a unified WAL-based realtime pipeline, establishing an open table format as the canonical analytics source and replacing independently managed sync processes
• Drive adoption of event-driven design best practices across engineering — eliminating anti-patterns, producing architectural guidance, and making correct consumption of the streaming platform the default
• Set the technical bar for the Realtime & Streaming team through design review, code review, and direct mentorship of senior engineers
• Participate in a 24/7 on-call rotation to provide continuous operational support and rapid incident response

Required Qualifications:

• 10+ years of production engineering experience, with deep specialization in realtime data infrastructure and event streaming systems
• Production experience designing and operating WAL-based or log-structured replication systems — PostgreSQL logical replication, Debezium, Kafka Connect, or equivalent CDC frameworks — at production scale, including delivery guarantees, consumer lag management, and failure recovery
• Deep understanding of event streaming correctness: ordering guarantees, at-least-once vs. exactly-once delivery semantics, idempotency requirements, and how those properties propagate to downstream consumers
• Experience leading batch-to-streaming migrations — replacing polling or batch-driven synchronization with event-driven architectures without disrupting dependent consumers mid-flight
• Fluency in a systems-level language — Rust, Go, or C/C++ — used to build high-throughput, low-latency data plane components in production
• A track record of leading cross-team technical initiatives — driving decisions through influence and producing documentation that holds up over time
• Strong written and verbal communication — you produce architectural guidance that the broader engineering organization depends on
• Comfortable operating with ambiguity — at this level, you will be handed problems, not plans, and are expected to identify the right approach and drive it from first principles to production
• Strong experience supporting 24/7 production on-call rotations, including responding to alerts and resolving incidents outside of standard business hours

(Optional) Preferred Qualifications:

• Systems-level programming experience in Rust for high-throughput streaming pipeline components — binary protocol handling, I/O-intensive services, or data plane infrastructure where performance and correctness are both hard constraints
• Experience with open table formats (Apache Iceberg, Delta Lake, Parquet) as the storage layer bridging streaming ingestion and analytical query workloads
• Hands-on experience integrating streaming pipelines with analytical query engines — Redshift, Snowflake, BigQuery — including schema evolution, backfill strategies, and latency and cost trade-offs
• Deep familiarity with PostgreSQL WAL internals — logical decoding, replication slots, and the failure modes that emerge at scale
• Familiarity with the streaming runtime landscape — Kafka, Kinesis, Redpanda, Pulsar, and similar — and the ability to evaluate trade-offs across managed and self-managed deployment models
• Experience with AWS data infrastructure — ECS, Kinesis, S3 — and how managed services compose with self-managed streaming components in a hybrid architecture
• Background in healthcare, pharmacy, financial services, or other regulated domains where data pipelines carry audit and correctness obligations

Nothing in this position description restricts management’s right to assign or reassign duties and responsibilities to this job at any time.