Database Architect

As a Data Platform / Database Architect specializing in Postgres and Kafka, you will play a crucial role in designing and optimizing a high-throughput, audit-friendly data platform supporting a SaaS for financial data automation and reconciliation. Your responsibilities will encompass owning the end-to-end design and performance of the data platform, including multitenant Postgres schemas, CDC pipelines, and analytics stores. Additionally, you will be instrumental in laying the groundwork for AI-powered product features. Key Responsibilities: - Design multitenant Postgres schemas with a focus on partitioning, indexing, normalization, and RLS, while defining retention and archival strategies. - Optimize Postgres performance by implementing strategies such as EXPLAIN/ANALYZE, connection pooling, vacuum/bloat control, query/index tuning, and replication. - Develop event streaming/CDC using Kafka/Debezium, including topics, partitions, schema registry setup, and data delivery to analytics engines like ClickHouse, Snowflake, and BigQuery. - Model analytics layers using star/snowflake schemas, orchestrate jobs with tools like Airflow/Dagster, and implement dbt-based transformations. - Establish observability and SLOs for data by monitoring query/queue metrics, implementing tracing, setting up alerting systems, and conducting capacity planning. - Implement data security measures including encryption, masking, tokenization of PII, and defining IAM boundaries to enhance the audit posture, such as PCI compliance. - Integrate AI components like vector embeddings (pgvector/Milvus), basic feature store patterns (Feast), retrieval pipelines, and metadata lineage. - Collaborate with backend, ML, and product teams to review designs, coach engineers, create documentation/runbooks, and lead migrations. Qualifications Required: - Minimum of 6 years of experience in building high-scale data platforms with in-depth expertise in PostgreSQL, covering areas like partitioning, advanced indexing, query planning, and replication/HA. - Hands-on experience with Kafka (or equivalent) and Debezium/CDC patterns, familiarity with schema registry setups (Avro/Protobuf), and understanding of exactly-once/at-least-once delivery semantics. - Proficiency in at least one analytics engine at scale such as ClickHouse, Snowflake, or BigQuery, along with strong SQL skills. - Working knowledge of Python for data tooling (e.g., pydantic, SQLAlchemy), experience in orchestration with Airflow or Dagster, and expertise in implementing transformations using dbt. - Solid experience in cloud environments like AWS/GCP/Azure, including networking, security groups/IAM, secrets management, and cost controls. - A pragmatic approach to performance engineering, excellent communication skills, and a knack for thorough documentation.,