Lead Agentic Data Systems Engineer

Salesforce•San Francisco, CA

1d•Remote

About The Position

Salesforce is seeking a hands-on, depth-first engineer for the role of Lead Agentic Data Systems Engineer. This individual will be responsible for transforming architectural blueprints into production-grade data products. The role involves end-to-end ownership, including building, maintaining, enhancing, and constructively challenging designs. The engineer will be the trusted party for making data products functional and high-quality. This role redefines the data team model by managing hand-off protocols between specialized AI agents, acting as the central anchor for a hybrid human-agent intelligence unit.

Requirements

Production-grade proficiency in Python, dbt, Airflow, and advanced SQL.
Expertise in Apache Spark, and Snowflake.
Fluency in AI-native development environments (e.g., Cursor, Codex, or Claude Code).
Expert in Prompt Engineering.
Mastery of agentic frameworks such as LangGraph.
Leverage MCP servers to retrieve data from tool stack.
Expert-level knowledge of chain-of-thought prompting, self-correction loops, and iterative reasoning paths.
Salesforce Core and Data 360 understanding.
Advanced understanding of Data Mesh, Data-as-a-Product (DaaP), and Event-Driven Architectures.
Understanding of Semantic layer and Knowledge Graphs.
Experience using agentic workloads via Docker, Kubernetes, and serverless compute environments.
5+ years of experience in high-stakes Data Engineering, Architecture, or Data Science.
A documented history of using generative AI to accelerate personal and departmental output by orders of magnitude.
The ability to function as a "Domain Data Officer," managing end-to-end data strategy for a business unit with minimal supervision.
Superior analytical judgment—the ability to identify subtle logic errors or hallucinations in agentic output before they reach production.

Responsibilities

Architect and maintain a private ecosystem of 10+ autonomous agents specialized in ETL, synthetic data generation, automated QA, and predictive modeling.
Design multi-step reasoning architectures and verification protocols to ensure agents autonomously validate and peer-review their own outputs.
Transform high-level, ambiguous business requirements into production-ready data products independently, bypassing the need for mid-level project management.
Use domain knowledge to ensure deployed tools are well governed, implementing governance as code for data pipelines and Agentic development.
Develop and maintain Model Context Protocol (MCP) servers to provide agents with secure, deep-link access to Snowflake, Salesforce, AWS, and proprietary internal data catalogs.