# Sr. Specialist Solutions Architect at Databricks

> Design and lead proofs of concept for production-grade AI agent and ML systems for enterprise customers.

- Company: Databricks (databricks.com)
- Location: Zürich, Switzerland
- Places: Zürich, Switzerland
- Posted: 2026-11-05
- Level: Senior
- Team: Specialist Solutions Architects – Switzerland, Austria & Eastern Europe
- Reports to: Manager, Specialist Solutions Architects – Switzerland, Austria & Eastern Europe
- Apply: https://databricks.com/company/careers/open-positions/job?gh_jid=8927382002
- Page: https://ye.saranatour.dev/jobs/sr-specialist-solutions-architect-databricks-kn7emeb2jqs599n0mp52n3j0458hmh7v

## The posting

Sr. Specialist Solutions Architect
Databricks · Zürich, Switzerland

FEQ228R125

As a Senior Specialist Solutions Architect (Sr. SSA) for AI/ML, you will be the go-to technical expert for Databricks' AI platform across our customers. You will help them move from AI experiments to **governed, cost-efficient agents and ML systems running on the Databricks Data Intelligence Platform**.

You will work with Account Executives and Solutions Architects to win strategic opportunities. You will design architectures for advanced agent and ML use cases and lead hands-on proofs of concept and MVPs. You will act as a trusted advisor to customer technical leaders on AI strategy, governance and economics. You will also shape our AI go-to-market in the region, enable the wider field team, and feed what you learn from customers back to Product and Engineering.

This role reports to the Manager, Specialist Solutions Architects – Switzerland, Austria & Eastern Europe.

### **The impact you will have**

- **Win the technical decision for AI on Databricks.** Lead discovery, architecture design, demos, proofs of concept and MVPs for agent, GenAI and ML use cases in competitive evaluations.
- **Design production-grade agent systems.** Architect agents and multi-agent applications using Agent Bricks, Databricks Apps, MCP tools, Lakebase, AI Search and the Foundation Model APIs. Govern assets under Unity Catalog
- **Make agent quality measurable.** Help customers build evaluation and observability into everything they ship, using MLflow tracing, domain-specific LLM judges, labelled evaluation datasets, release gates and production monitoring.
- **Govern AI at enterprise scale.** Position the AI Gateway as the central control point for model access, guardrails, PII protection, usage tracking and rate limiting.
- **Scale your expertise.** Enable SAs and partners with workshops, reusable assets and reference architectures.
- **Influence the product.** Bring structured customer feedback and requirements specific to the market to Product Management and Engineering.
- **Be a visible voice.** Represent Databricks at customer executive briefings, meetups, conferences and in written content.

### **What we look for**

- 7+ years of hands-on experience in machine learning, AI engineering or data science, including customer-facing technical work (pre-sales, consulting or technical leadership).
- **A track record of shipping LLM and agent systems to production**, not just prototypes, including evaluation, monitoring and cost management.
- Deep knowledge of agent architecture: tool use and function calling, MCP, agent memory and state, orchestrating multiple agents, and retrieval and context design.
- Strong practical experience **evaluating GenAI systems**: tracing, LLM-as-a-judge, building evaluation datasets, regression testing and observability in production.
- Hands-on with at least one modern agent framework (e.g. LangGraph, OpenAI Agents SDK, Google ADK, DSPy). You should be framework-agnostic in how you design.
- Solid understanding of the foundation-model landscape (e.g. Anthropic, OpenAI, Google, open-weight models) and the trade-offs between quality, latency, cost and data residency.
- Strong grounding in classical ML and MLOps: model development, deployment, monitoring and lifecycle management at scale.
- Strong software engineering skills in Python and SQL. You use AI coding assistants (e.g. Claude Code, Cursor) daily and can build a working demo application qu