Senior Software Engineer

Microsoft Microsoft · Big Tech · Vancouver, BC +1 · Software Engineering

Senior Software Engineer role focused on building LLM-powered data engineering experiences and infrastructure for Microsoft Fabric, utilizing Apache Spark. The role involves implementing agentic workflows, scalable LLM-backed data features, and defining evaluation strategies for LLM features, while applying Responsible AI practices.

What you'd actually do

  1. Design, build, and ship scalable backend services and/or libraries in Python that power Fabric Data Engineering and Data Science experiences
  2. Develop LLM-enabled capabilities (prompting patterns, tool/function calling, RAG/grounding, orchestration/agents) with strong attention to latency, reliability, and cost
  3. Build robust data pipelines and distributed compute solutions (Spark/PySpark) to support model/data workflows, feature generation, and large-scale analytics
  4. Define evaluation strategies for LLM features (offline/online metrics, quality gates, safety checks), and implement telemetry/monitoring to continuously improve quality
  5. Apply Responsible AI and security/privacy best practices (data handling, governance, access controls) when integrating AI into customer-facing products

Skills

Required

  • 4+ years technical engineering experience with coding in languages including, but not limited to, C, C++, C#, Java, JavaScript, or Python
  • 4+ years experience in Frontend + UX engineering skills: React + TypeScript, accessibility, performance, and building user-centered flows.
  • 4+ years experience in Backend / full-stack fundamentals: service/API design, debugging distributed systems, reliability/operability, and production ownership.

Nice to have

  • Experience building and operating cloud services (Azure preferred), including telemetry, monitoring, experimentation/rollout strategies, and cost/latency awareness.
  • Experience with data engineering concepts and systems (e.g., Spark, notebooks, lakehouse-style workflows) and the needs of professional data engineers.
  • Understanding of modern AI/LLM-assisted product patterns (tool use, grounding, evaluation mindset, trust/safety guardrails) and how to ship these experiences.
  • Ability to collaborate across disciplines (PM, Design, Research, partner engineering teams) and drive ambiguous problems to crisp execution.

What the JD emphasized

  • LLM-powered data engineering experiences
  • agentic workflows
  • scalable LLM-backed data features
  • evaluation strategies for LLM features
  • Responsible AI

Other signals

  • LLM-powered data engineering experiences
  • agentic workflows
  • scalable LLM-backed data features
  • evaluation strategies for LLM features
  • Responsible AI