Senior Code Search & Retrieval Engineer

Mercor; San Francisco (CA); On-site; USD 180,000 - 260,000; Full time

Job summary

Mercor is hiring a Software Engineer for Code Search & Retrieval in San Francisco. You will own end-to-end architecture for code search, combining dense code embeddings with BM25, and determine when to route tasks to specialized models.

The role blends research, systems engineering, and hands-on coding to deliver scalable, accurate code retrieval. You will mentor engineers, drive design reviews, and shape long-term strategy for code retrieval, embeddings, and model selection as SOTA advances are

Qualifications

• 8+ years of professional software engineering experience, including 3+ years at Senior level or above with staff-level impact.
• Deep, hands-on expertise building search and retrieval systems: dense-embedding retrieval, lexical scoring (BM25/TF-IDF), hybrid ranking and re-ranking.
• Experience with code search or code understanding is a plus but not required.
• Strong understanding of search algorithms and index internals: vector/ANN indices, inverted indices and engines such as Elasticsearch/OpenSearch, Lucene, FAISS, or vector databases.
• Track record of making cost-vs-speed tradeoffs: latency budgets, throughput, memory footprint, and infra spend on high-QPS systems.
• Familiarity translating natural language queries into effective search queries.

Responsibilities

• Own the architecture of Mercor's code search and retrieval systems—hybrid retrieval with dense code embeddings and BM25.
• Identify and rank similar code tasks and route tasks to the right models.
• Design natural-language-to-query translation for code search.
• Build the indexing pipeline to keep the task index fresh as new tasks and results arrive.
• Balance cost and speed via embeddings, quantization, ANN indices, caching, and sharding.
• Prototype systems, ship production code, and unblock teams on retrieval and infrastructure problems.
• Mentor engineers and lead design reviews across the org.
• Collaborate with product, researchers, and leadership on platform investments and hiring.

Skills

8+ years software engineering

Senior level / Staff-level impact

Search and retrieval systems

Cost-vs-speed tradeoffs

Mentoring engineers

Prototype and shipping production code

Natural language to search translation

Tools

Elasticsearch/OpenSearch

Lucene

FAISS

Vector databases

HNSW

Job description

Mercor is hiring a Software Engineer for Code Search & Retrieval in San Francisco. You will own end-to-end architecture for code search, combining dense code embeddings with BM25, and determine when to route tasks to specialized models.

The role blends research, systems engineering, and hands-on coding to deliver scalable, accurate code retrieval. You will mentor engineers, drive design reviews, and shape long-term strategy for code retrieval, embeddings, and model selection as SOTA advances are