Fullstack AI Software Engineer (Agentic Systems, Fullstack Builder)

At Sonder, we believe that every person deserves to feel safe, supported, and empowered to be at their best - wherever they are. Sonder's mobile platform provides 24/7, real-time support from a dedicated team of safety, medical, and mental health professionals.

About the Role
Our Customer Insights and Measurement team is looking for a Full Stack Engineer to help us build the next generation of wellbeing intelligence. We are looking for a builder who wants to own the entire journey, from crafting the customer-facing experience and data models to architecting the AI-first, human-in-the-loop insights that power our platform.

You will own the end-to-end journey of turning raw data into intelligent insights. On any given week, you might be building React-based dashboards, optimizing Snowflake transformations, or architecting multi-agent workflows that help our customers "talk" to their safety data.

What You'll Be Doing
Full Stack Development

• End-to-End Ownership: Build and maintain features from the database schema and Snowflake models through to the React UI.
• API Architecture: Design robust back-end services (Python/Node.js) that support data ingestion and high-concurrency reporting.

AI & Agentic Systems

• Agentic Workflows: Design and ship autonomous AI workflows using frameworks like LangGraph, CrewAI, or native tool-use patterns.
• RAG & Context Engineering: Build and optimize RAG pipelines, focusing on retrieval quality, chunking strategies, and metadata filtering.
• MCP Integration: Build and deploy Model Context Protocol (MCP) servers to bridge the gap between our internal data silos and LLM reasoning engines.
• AI Observability: Implement evaluation frameworks (e.g., LangSmith, Arize Phoenix) to monitor "vibe checks," latency, and cost in production.

Data Pipelines & Reporting

• Snowflake Mastery: Design and optimize tables and views that power customer-facing dashboards.
• Insight Automation: Use LLMs to automate the synthesis of complex safety data into actionable executive summaries.

What You Will Bring

Must-Have:

• The "Builder" DNA: heavy user of AI-assisted coding tools (Cursor, ClaudeCode, or Copilot) shipping production AI.
• Full Stack Fluency: Deep experience in React/Next.js and Python (preferred for AI) or Node.js.
• Data Engineering Literacy: Strong SQL skills and experience with data warehousing (Snowflake strongly preferred).
• Agentic Experience: Understand the difference between a basic prompt and a multi-step agentic loop involving tool-calling and structured outputs.
• Production Mindset: Care about LLM evaluation, prompt versioning, and error handling.

Nice-to-Have: vector databases (Pinecone, pgvector), dbt, LangChain/LangGraph orchestration frameworks, AI-first side projects.