Principal Data Architect

Easyberry; San Francisco (CA); On-site; USD 140,000 - 190,000; Full time

Job summary

Easyberry is seeking a senior data engineer to own the canonical data model and master data management for enterprise-scale AI initiatives. You will design a data platform in Microsoft Fabric, align data across divisions, and govern data quality and semantics for AI readiness.

You will mentor junior engineers, communicate with stakeholders, and balance rapid delivery with solid architecture in a small, highly capable team.

Qualifications

• 10+ years in data engineering and architecture with senior or principal ownership.
• Deep data modeling expertise across conceptual/logical/physical layers.
• Hands-on Microsoft Fabric experience or deep Azure data platform skills.
• Ability to handle enterprise-scale data and post-acquisition integrations.
• Ability to translate complex concepts for stakeholders and drive decisions.

Responsibilities

• Define the canonical data model across business units.
• Lead master data management: golden records, survivorship rules, stewardship workflows.
• Design data platform architecture in Microsoft Fabric, including lakes, layers, and pipelines.
• Ensure data quality, governance, and accountability after project completion.
• Build data foundations for AI and ensure semantic layers support agents.

Skills

Senior data engineering

Data modeling

Ontology design

Communication with stakeholders

Problem solving

Tools

Microsoft Fabric

Azure data tools (Synapse, Data Factory)

Databricks

Purview

Job description

We're a forward deployed engineering studio. We embed with enterprise clients to help them figure out where AI actually belongs in their business, then build the systems that get them there. That means real problem solving and architecture work up front, and a paved path our clients' own teams can follow once we've built it.

You'd be working alongside partners who've built at top companies like Meta, Amazon, and Airbnb, and who've founded venture-backed startups. Our clients are large enterprises with billions in revenue and real operational complexity.

The role

About us

We're a forward deployed engineering studio. We embed with enterprise clients to help them figure out where AI actually belongs in their business, then build the systems that get them there. That means real problem solving and architecture work up front, and a paved path our clients' own teams can follow once we've built it.

You'd be working alongside partners who've built at top companies like Meta, Amazon, and Airbnb, and who've founded venture-backed startups. Our clients are large enterprises with billions in revenue and real operational complexity.

AI is only as good as the data underneath it. Our clients have decades of operational data spread across divisions, systems, and acquisitions. You'd own the work of turning that into a coherent, well-governed foundation.

You'd be the senior data voice on our engagements: designing the ontology, standing up the platform, and getting an organization to actually agree on what its data means.

This is a small team. We need someone who can take an ambiguous problem and run with it.

What you'll own

• The canonical data model. Defining core entities, relationships, and the domain ontology across business units that currently disagree
• Master data management: golden records, match and merge logic, survivorship rules, stewardship workflows
• Data platform architecture in Microsoft Fabric. Lakehouse design, medallion layers, pipelines, and the cost and performance tradeoffs behind them
• Data quality and governance. What gets measured, who owns it, how it stays true after we leave
• Making the data usable for AI. Grounding, retrieval, and the semantic layer that lets agents and copilots actually answer questions correctly
• Scoping honestly. Telling us which domains are worth tackling first and which ones aren't ready
• Mentoring and raising the bar on more junior data engineers
  

What we're looking for

Required

• 10+ years in data engineering and architecture, with senior or principal-level ownership
• Deep data modeling expertise. Conceptual, logical, and physical, and you have opinions about dimensional versus data vault versus graph and when each earns its place
• Hands-on Microsoft Fabric experience (or strong Azure data platform depth: Synapse, Data Factory, Databricks, Purview)
• Enterprise scale. Messy source systems, legacy platforms, post-acquisition integration
• Ability to break down complex technical concepts into simple terms so stakeholders can understand
• Facilitation skill. Much of this job is getting people who disagree about definitions into a room and walking out with a decision

Nice to have

• Experience in both startup and large company environments, so you know when to move fast and when process actually earns its keep
• Enterprise or consulting delivery experience, comfortable in front of a client
• Building data foundations specifically for AI and LLM use cases
• Knowledge graphs, semantic layers, or formal ontology work
• Data governance and privacy at regulated or enterprise scale
• Complex operational domains: construction, energy, manufacturing, logistics, or field services
  

How we work

Small team, direct communication, short feedback loops. Two-week sprints. You'll work closely with engineering, product, and design, and you'll have real say in what we take on and how we scope it. If you think something is a bad idea, we want to hear it early.

Why this is interesting

Most enterprise AI stalls on data, not models. The pilots work in a demo and fall apart the moment they hit real systems. You'd be doing the unglamorous foundational work that decides whether any of it graduates, at clients that are committed, well-resourced, and moving fast. It's a rare chance to design an ontology from the ground up at real scale, and then actually build it.