← Workday Interview Insights

Workday·Software Engineer·Onsite - System Design / Architecture·Senior

Senior
May 2026

Summary

Workday system design round, one big question about building a document upload and search platform with an AI extraction pipeline in the middle. Pretty dense for a single interview slot, they wanted the full picture from blob storage all the way to search ranking.

Questions Asked (1)

Q1

Design a system where users can upload documents and search them using both structured field queries and free-text keywords, with field extraction handled by a multi-step AI pipeline.

System DesignTechnical Trade-offsAPI & Integrations
Author's notes

This was a lot to cover in one question.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clarifying requirements and scale, then propose a high-level architecture that separates document ingestion, AI-powered field extraction, and a hybrid search layer. Emphasize trade-offs in the AI pipeline (accuracy vs. latency/cost) and how you'd handle failures and reprocessing. Conclude with how you'd expose the search via APIs and ensure consistency between extracted fields and free-text indexes.

Pro tip: Show awareness that AI extraction is probabilistic: design for confidence scores, human-in-the-loop review for low-confidence fields, and idempotent reprocessing when models improve. This demonstrates production maturity beyond just wiring up an LLM.

1. Clarify Requirements and Scale

Ask about document types, expected volume, latency requirements, and search expectations. Establish whether extraction is synchronous or asynchronous and what accuracy is acceptable.

2. Design Ingestion and Storage

Outline a pipeline where uploaded documents are stored in object storage, metadata in a relational DB, and raw text in a search index. Use a queue to decouple upload from processing.

3. Define the Multi-Step AI Extraction Pipeline

Break extraction into stages: OCR/parsing, entity recognition, field mapping, and validation. Discuss using LLMs or specialized models per stage, with confidence scores and fallback logic.

4. Implement Hybrid Search

Combine structured field queries (e.g., SQL or Elasticsearch filters) with free-text search (e.g., BM25 or vector search). Explain how extracted fields are indexed and kept in sync.

5. Address Trade-offs, Failures, and Evolution

Discuss latency vs. accuracy, cost of AI calls, handling extraction failures, and reprocessing documents when models improve. Mention monitoring and human review for low-confidence extractions.

Key Points to Mention

  • Asynchronous processing with message queues to handle upload spikes and decouple AI pipeline
  • Confidence scores and human-in-the-loop validation for extracted fields
  • Hybrid search architecture combining structured filters and free-text (e.g., Elasticsearch with both keyword and text fields)
  • Idempotent reprocessing and versioning of extraction models to allow re-indexing without data loss
  • API design for upload, search, and status endpoints, including pagination and error handling
  • Trade-offs between using a single large LLM vs. a multi-step pipeline of smaller models (cost, latency, accuracy)

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.