Principal Knowledge Acquisition Analyst

Autodesk Autodesk · Enterprise · Poland · Remote

Autodesk is seeking a Principal Knowledge Engineer to design and operate AI-powered systems for capturing and transforming knowledge from enterprise engagements into structured, reusable guidance. This role involves building AI-assisted pipelines, optimizing AI agents, defining quality metrics, and ensuring data is ready for downstream content production and AI use cases like RAG.

What you'd actually do

  1. Design and operate AI-powered pipelines to capture knowledge from enterprise systems, meeting transcripts, and engagement artifacts
  2. Manage and optimize AI agents, including prompt design, evaluation, and performance tuning
  3. Define and track quality metrics (accuracy, completeness, error rates) and continuously improve pipeline performance
  4. Design and operate pipelines that convert raw, unstructured inputs into structured, template-aligned outputs using AI
  5. Ensure structured outputs support AI-driven use cases such as retrieval (RAG) and downstream content generation

Skills

Required

  • 8+ years of experience in knowledge management, data engineering, information systems, or related fields
  • Experience working with AI/LLM-based workflows in production
  • Experience designing data pipelines, structured capture, or transformation processes
  • Strong analytical skills with ability to define and improve quality metrics
  • Experience working cross-functionally with product, engineering, and domain experts

Nice to have

  • Knowledge management platforms: Confluence, SharePoint, Notion Enterprise, Gainsight Knowledge, Guru
  • AI / LLM : OpenAI / AzureOpenAI , Anthropic Claude, Google Gemini
  • Agentic Workflow & Orchestration: ReAct (Reason + Act), Chain of Thought (CoT) patterns
  • AI Operations: Prompt design and evaluation, LLM output evaluation and benchmarking, Model Monitoring and QA, Vector Databases, Markdown files
  • Cloud environments: Azure, AWS, Google
  • Has built AI-driven knowledge capture or data pipelines at scale
  • Can transform messy, real-world data into structured, usable outputs
  • Thinks systemically about data, content, and downstream use
  • Works effectively across technical and non-technical teams
  • Focuses on building practical, scalable systems

What the JD emphasized

  • AI/LLM-based workflows in production
  • built AI-driven knowledge capture or data pipelines at scale

Other signals

  • designing and operating AI-powered pipelines
  • building AI-powered extraction and transformation pipelines
  • establish scalable AI-assisted capture and transformation pipelines