Description
SoFi's ability to deliver trusted member experiences, manage risk, meet regulatory obligations, and scale enterprise AI depends on data that is reliable, governed, understandable, and available when it is needed.
As a Principal Software Engineer in the Data Platform Group, you will provide hands-on technical leadership within the Data Trust pillar.
Your work will span three connected areas: establishing shared Data Trust capabilities, raising the reliability of data across batch and low-latency environments, and applying those foundations to a semantic context layer for trusted analytics and AI.
By making business meaning, ownership, provenance, quality, freshness, and access policies machine-readable, you will help end users and AI systems work from consistent, governed context.
Responsibilities:
- Data Trust architecture and strategy: Define and execute the technical direction for data contracts, metadata and catalog modernization, lineage, data quality, privacy, and access controls.
- Declarative enforcement primitives: Build platform services, libraries, and reference patterns that make contracts, quality rules, privacy policies, and access requirements versioned, testable, observable, and automatically enforceable across the data lifecycle.
- Semantic context for analytics and AI: Establish the technical foundations for a governed semantic context layer that connects business definitions, metadata, lineage, ownership, quality, freshness, and access policy.
- Reliability across batch and streaming data: Lead the evolution of Data Trust capabilities across Snowflake, AWS data services, Kafka-based ingestion, and other batch or low-latency environments.
- Mission-critical operational rigor: Raise the standard for production data systems through observability, integrity checks, audit evidence, incident learning, and safe automation.
- Cross-organizational technical leadership: Drive multi-quarter initiatives across teams and functions, earning alignment on difficult architectural tradeoffs and investment decisions.
What success looks like in the first 12–18 months:
- A clear enterprise architecture and delivery path exists for Data Trust primitives spanning contracts, metadata, lineage, quality, privacy, and reliability.
- Priority data domains have materially improved coverage, observability, SLOs, and automated enforcement.
- The platform provides production-ready patterns for governing data in both batch and low-latency environments.
- Initial semantic context capabilities are operating in production and provide more reliable, traceable context for analytics and AI use cases.
- Manual investigation, recurring data failures, and compliance evidence-gathering are measurably reduced.
- Engineering and partner teams have adopted reusable standards and capabilities rather than creating isolated point solutions.
Requirements:
- 15+ years of software or data engineering experience (or 13+ years with a Master’s degree) with a proven track record of designing, building, and operating distributed data platforms at enterprise scale.
- Deep experience designing horizontal, multi-tenant platform services that other engineering teams depend on.
- Practical, battle-tested experience building or managing components of the modern data trust stack, such as data contracts, metadata catalogs, data quality frameworks, end-to-end lineage, and fine-grained access control (RBAC/ABAC).
- Advanced knowledge of modern data ecosystems, specifically Snowflake, Kafka (or similar streaming platforms), and AWS data services.
- Mastery of one or more core programming languages (e.g., Python, Java, Go) and advanced SQL.
- A history of successfully driving multi-team, cross-functional technical initiatives.
- Experience managing Tier-0/mission-critical infrastructure.
Nice to have:
- Prior experience operating in a highly regulated environment where auditability, data privacy (CCPA/GDPR), and data integrity are architectural requirements, not afterthoughts.
- Experience building semantic layers, knowledge graphs, or context engines that ground LLMs and autonomous agents in governed enterprise data.
- Contributions to relevant open-source data or metadata projects (e.g., Apache Iceberg, OpenMetadata, DataHub, Amundsen).
Compensation and Benefits:
The base pay range for this role is listed below. Final base pay offer will be determined based on individual factors such as the candidate’s experience, skills, and location.