Distinguished Architect, Data Platform

CloudZero•San Francisco, CA

18h

About The Position

CloudZero is growing fast. Our customer base is expanding, the data challenges we're solving are getting more complex, and the platform is scaling to match. As a Distinguished Architect on the Data Engineering team, you'll own some of the hardest infrastructure problems at CloudZero: shaping the next-generation streaming data platform, the dimensional cost model underlying every attribution decision, the hot/cold storage architecture serving both real-time and historical queries, and the query engine that powers our entire product. This is real platform architecture work at real scale, not a consulting role or a review-and-advise job. You'll define the roadmap, drive the foundational decisions, and be a force multiplier for a talented engineering team — evolving CloudZero from batch-oriented pipelines toward a streaming-first architecture where cost attribution reaches engineers within seconds of a resource being used, not the next morning. This role is ideal for an architect who has built systems like this before, has the scars to prove it, and wants to see their decisions matter in direct and measurable ways for customers and for the business.

Requirements

10+ years in data engineering with a clear trajectory toward principal or staff-level architecture
Built and operated large-scale data platforms serving tens of millions of events per day in production
Deep experience with streaming systems such as Kafka, Kinesis, Flink, or Spark Streaming at real production throughput
Strong hands-on fluency with modern open table formats including Apache Iceberg, Delta Lake, and Hudi, including compaction, partitioning strategy, and time-travel queries
Designed hot/cold storage architectures with explicit latency SLOs per tier
Proven ability to drive a data platform end to end, not just a single layer
Expert in dimensional data modeling including fact/dimension schema design, slowly changing dimensions, and cardinality management
Deep understanding of the materialization tradeoff space: full vs. incremental, push vs. pull, pre-aggregate vs. query-time
Experience with cost attribution, showback/chargeback, or multi-tenant data partitioning patterns
Strong SQL and query optimization background across predicate pushdown, partition pruning, and cost-based query planning
Hands-on with distributed query engines such as Trino, Presto, Spark SQL, or DuckDB including configuration, optimization, and production operations
Understands catalog and metadata management and how it couples to query engines
Comfortable with cloud data warehouses such as Snowflake, BigQuery, and Redshift and how they integrate with open table formats
Experience driving query engine migrations while maintaining production SLAs
Track record as a technical anchor for a data platform or data engineering team
Writes clear ADRs, RFCs, and technical design docs that bring engineers along
Can drive multi-month, multi-team technical initiatives from inception to production without heavy process overhead
Communicates complex tradeoffs to non-technical stakeholders including product and business leadership
Comfortable in a high-autonomy environment: builds consensus, influences through expertise, and helps teams move forward

Nice To Haves

FinOps or cloud cost domain experience
Multi-cloud data ingestion across AWS, Azure, and GCP
Apache Flink at production scale
Lakehouse architecture patterns
Real-time feature engineering for ML
Data mesh or domain-oriented design patterns
Prior startup or high-growth SaaS experience
Open source contributions to the data ecosystem

Responsibilities

Define the Data Platform Architecture
Lead end-to-end technical design for CloudZero's next-generation data platform, from event ingestion and stream processing through hot/cold storage and the query layer to the API surface
Document architectural decisions, tradeoffs, and migration strategies with the rigor of an RFC-driven process
Shape and drive every layer of the new architecture: event ingestion, stream processing and enrichment, real-time serving, analytical storage, query layer, and API
Drive Streaming Infrastructure to Production
Design and deliver CloudZero's real-time data pipeline from ingestion through enrichment to serving
Establish SLOs for throughput, latency, and correctness, and build the operational playbooks that make this system trustworthy enough to replace the batch pipelines our entire product currently depends on
Tackle real-time streaming at scale across thousands of customers simultaneously, with fault tolerance, backpressure awareness, and correctness as non-negotiables
Tackle the Dimension Cardinality Problem
Redesign CloudZero's dimensional cost model to support high-cardinality, multi-dimensional cost attribution without runaway materialization costs
Drive incremental, delta-based materialization strategies using modern open table formats, dramatically reducing expensive full-rebuild jobs and unlocking millions in annual infrastructure savings
Evolve the Query Layer
Assess CloudZero's current query infrastructure, drive in-flight migrations to completion, and lead the evolution of the query engine layer going forward
Own performance optimization across partition pruning, predicate pushdown, and query planning, and set the vision for how the query layer grows as data volumes scale 10x
Extend Cost Attribution to Real-Time
Evolve CloudZero's proprietary cost attribution engine from a batch-oriented model to one that assigns complex cost dimensions by team, feature, and customer within seconds of resource usage
Rethink enrichment, data lineage, and correctness guarantees in a streaming context
Shape the Data Engineering Roadmap
Partner with product, infrastructure, and analytics engineering to define a multi-year data platform roadmap
Build consensus across engineering leadership on foundational investments including table formats, streaming frameworks, query engines, and schema management
Elevate the Engineering Team
Participate in architecture reviews, contribute to design patterns and best practices, and mentor senior and staff engineers through code review, pairing, and structured feedback
Make everyone around you better, not by directing, but by raising the collective craft