AI Data Engineer

About Port

Port is the Agentic SDLC Platform. The way engineering teams build software is constantly evolving, and every new wave - such as AI agents operating across the SDLC- makes engineering more powerful, but also more complex and chaotic.

Port turns that chaos into control: one platform to build, govern, and operate the software development lifecycle, enabling every team to move as fast as technology allows.

Following our $100M Series C funding round, we are entering an exciting stage of significant growth and global expansion. As a product-led company, we believe the recipe for success is a world-class product that speaks for itself.

About The Role

We're building the intelligence layer that lets everyone at Port ask hard questions of our data and get trustworthy answers — in plain language, in seconds. As AI Data Engineer, you own the semantic and AI-native surface of our Snowflake platform: the governed semantic layer that defines "what a metric means," the Cortex Agents that let business users query it conversationally, and the infrastructure that keeps all of it fast, reliable, and ready to scale.

This is a builder-owner role. You'll ship the semantic models, wire up the agents, and set the standard for how analysts across the company work with data. You'll also lead the Data Analyst guild — the connective tissue that keeps our hub-and-spoke analytics model coherent as it grows.

If you're excited by the space where data engineering, LLM tooling, and analytics governance meet, this role sits right in the middle of it.

What you’ll Own:

• Snowflake semantic layer — Own the semantic layer end-to-end as the single source of truth for metrics; design semantic models, enforce naming standards, and ensure consistent metric definitions across dashboards and AI agents.
• Cortex Agents — Design and deploy conversational AI agents using Cortex Analyst and Cortex Search; tune for accuracy and safety, expose through multiple surfaces (Snowflake Intelligence, Streamlit, MCP), and build evaluation harnesses to maintain quality at scale.
• Data craft (Analytics guild) — Co-lead the technical track of the Analytics guild; set SQL and modeling standards, run technical enablement and code reviews, and serve as the technical authority for analysts.
• Scaling data infrastructure — Improve performance, reliability, cost efficiency, and governance across the dbt/Airflow/Airbyte/Snowflake stack as data volume and query load grow; optimize warehouse sizing, medallion layers, and ingestion pipelines.
  
  

• Model and maintain semantic views that power both dashboards and AI agents, keeping definitions versioned, tested, and certified.
• Build, evaluate, and iterate on Cortex Agents — including retrieval quality, guardrails, and observability.
• Extend and optimize dbt models, Airflow DAGs, and Airbyte connectors across bronze/silver/gold layers.
• Partner with GTM, Finance, CS, and Product stakeholders to translate business questions into governed, reusable data assets.
• Run the Data Analyst guild: standards, reviews, enablement, and tooling.
• Own data quality, lineage, and cost monitoring across the Snowflake platform.
• Expose data and agents through Streamlit apps, BI tools (Omni), and MCP servers for internal AI workflows.
  
  

• 5+ years of experience in data engineering roles in B2B SaaS companies
• Strong SQL and hands-on data engineering experience building production pipelines (dbt strongly preferred; orchestration with Airflow or similar).
• Deep, practical Snowflake experience — warehouse management, performance tuning, cost control, and data modeling.
• Experience building or maintaining a semantic / metrics layer, and a strong point of view on metric governance.
• Hands-on work with LLM-powered data applications — RAG, text-to-SQL, agent orchestration, or similar. Snowflake Cortex (Analyst, Search, Agents) is a big plus.
• A builder mindset paired with the judgment to set standards others follow.
  
  

• Experience leading a guild, chapter, or community of practice — or otherwise driving standards without direct authority.
• Familiarity with reverse-ETL, streaming ingestion (Airbyte or similar), and BI tooling on a semantic layer (Omni, ThoughtSpot).
• Exposure to GTM / RevOps data (CRM, product usage, call intelligence) and the ambiguity that comes with it.
• Experience with MCP, Streamlit, or embedding AI into internal tooling.