Mercor
Software Engineer, Search Systems - Code Data
About this role
Lead the design and implementation of Mercor's code search and retrieval systems, combining dense embeddings and lexical signals to find similar code tasks at scale. This hands-on Staff+ role focuses on solving complex code-specific retrieval problems, routing tasks to appropriate models, and continuously evolving search quality as frontier AI models advance.
What you'll do
- Design and own end-to-end hybrid retrieval architecture combining code embeddings, BM25, ranking, and re-ranking for code tasks
- Build systems to identify semantically similar code tasks beyond surface-level text matching, capturing structure, intent, and difficulty
- Create code-specific model selection and routing systems to match tasks with appropriate models
- Develop natural-language-to-query translation for precise code and task search
- Design and operate indexing pipelines balancing real-time ingestion, incremental updates, and consistency as new tasks arrive
- Define evaluation metrics, establish testing frameworks, and mentor engineers across the organization
What they're looking for
- Information retrieval and search systems architecture
- Dense and sparse embeddings (code embeddings, BM25)
- Approximate nearest neighbor (ANN) indexing and vector databases
- Large language models and code-specific models
- Distributed systems and indexing pipeline design
- Cost and performance optimization at scale
- A/B testing and online/offline evaluation frameworks
- Technical leadership and mentorship
Opens the application — the Jobs AI extension fills it for you. Set up autofill
Opens the official application on the employer’s site. No login required.
Mercor
Mercor builds a marketplace platform connecting expert talent to AI opportunities, supported by identity infrastructure, matching algorithms, and internal tools for data management. The company is hiring Software Engineers, Machine Learning Engineers, Fullstack Engineers, and Security Engineers to develop backend systems, ML models, cloud infrastructure, and distributed platforms.
- Website
- mercor.io
Likely interview questions
- Walk us through how you'd design a hybrid retrieval system combining dense code embeddings and BM25—what are the key tradeoffs?
- How would you approach identifying semantic similarity in code tasks beyond keyword matching?