Sr. Principal AI Systems Architect
AI summary of the role
Fivetran is hiring a Sr. Principal AI Systems Architect to own the architecture behind the analyst AI system, including metadata contracts, tool interfaces, context assembly, prompt and skill orchestration, evaluation loops, and operational patterns.
What you’ll do
- Define the architecture for the analyst AI stack across prompts, tools, metadata, memory and context strategies, evaluation, and runtime safety.
- Build the abstractions and interfaces that allow product teams to add new agent capabilities without creating brittle prompt sprawl.
- Design retrieval and context strategies that give the model the right information at the right time with strong cost, latency, and quality tradeoffs.
- Establish evaluation loops, benchmark suites, and review processes so model behavior can be measured, debugged, and improved systematically.
What you’ll bring
- Deep experience building AI-enabled product systems, not just running experiments.
- Strong software architecture skills, especially around APIs, contracts, orchestration, and system boundaries.
- Practical understanding of LLM prompting, tool use, retrieval, structured outputs, and evaluation methods.
- Ability to work fluently across backend engineering, product constraints, and user workflow design.
Technologies
LLM · prompting · retrieval · structured outputs · APIs · orchestration · metadata contracts · evaluation loops · benchmark suites
About Fivetran
Fivetran automates governed data movement from SaaS apps, databases, ERPs, and files into warehouses/lakes, then transforms and activates it for analytics, operations, and AI.
Private Late · 1000–2000 people
Source and classification
Internal deployment & tooling · Evidence for this classification:
extensible. This role owns the architecture behind the analyst AI system, including metadata contracts, tool interfaces, context assembly, prompt and skill orchestration, evaluation loops, and the operational patterns that make agents useful in real workflows instead of isolated demos. This is a systems role as much as an AI role. The challenge is not just getting an LLM to do something impressive once. The challenge is building a product-quality analyst system that can reason over schema and business context, use tools correctly, recover from ambiguity, and improve through disciplined evaluation and iteration. This is a full-time, hybrid position based at our Oakland, CA office. Our hybrid work model offers a blend of remote flexibility and in-person collaboration, including two days in the office each week to connect and build as a team. What You'll Do Define the architecture for
More from the job description
From Fivetran’s founding until now, our mission has remained the same: to make access to data as simple and reliable as electricity. With Fivetran, customer data arrives in their warehouses, canonical and ready to query, with no engineering or maintenance required. We’re proud that more organizations continue to leverage our technology every day to become truly data-driven. About Us Fivetran and dbt Labs are bringing together two industry-leading companies with a shared mission: helping organizations unlock the full value of their data. Together, we’re delivering the data infrastructure layer that helps organizations move, transform, and trust their data — from the moment data moves, through every transformation, to the context teams and AI systems rely on. Fivetran helps organizations automate data movement across the systems, clouds, engines, and tools they rely on. dbt Labs pioneered analytics engineering, helping teams transform data into reliable, governed insights. Together, we support thousands of organizations as they build a trusted foundation for analytics, AI, and better business decisions. As we bring our teams and technology together, we’re building on the strengths of both companies while continuing to deliver the products and experiences our customers know and trust. It’s an exciting time to join us: we’re creating a company with the scale, talent, and technolo [... source excerpt omitted ...] CA office. Our hybrid work model offers a blend of remote flexibility and in-person collaboration, including two days in the office each week to connect and build as a team. What You'll Do Define the architecture for the analyst AI stack across prompts, tools, metadata, memory and context strategies, evaluation, and runtime safety. Build the abstractions and interfaces that allow product teams to add new agent capabilities without creating brittle prompt sprawl. Design retrieval and context strategies that give the model the right information at the right time with strong cost, latency, and quality tradeoffs. Establish evaluation loops, benchmark suites, and review process [... source excerpt omitted ...] g, and One Team, One Dream. Read about us in Forbes. Fivetran brings together high-quality talent across the globe to make data access as easy and reliable as electricity for our customers. We value and recognize that our customers benefit from having innovative teams made of people from many backgrounds, experiences, and identities. Fivetran promotes diversity, equity, inclusion & belonging through attracting, recruiting, developing, and retaining a diverse workforce, not only because it is the right thing to do, but because it helps us build a world-class company to better serve our customers, our people and our communities. To learn more about Fivetran’s culture and what it’
Employer postings · Data from · Sources