Agentic Search Engineer: Multimodal Data & AI

ByteDance; Seattle (WA); On-site; USD 154,000 - 301,000; Full time

Job summary

ByteDance in the United States is seeking engineers for the Search team to build Agentic Search and analytics platforms. You will develop multi-source, multimodal search across structured data, documents, images, audio, and video, and unify enterprise data with a semantic layer of entities and relationships.

You will work with LLMs, agents, RAG, and vector databases to enable AI-driven information retrieval at scale.

Qualifications

• Degree in CS/AI/ML or related field.
• Strong coding and data structures knowledge with Python/Java/C/C++/Golang.
• Familiarity with LLMs, Agents, Search, RAG, or ML.
• Strong problem-solving and communication skills.

Responsibilities

• Build Agentic Search and Analytics systems combining planning, retrieval, reasoning, tool execution, and loops.
• Develop multi-source and multimodal search across data, documents, images, audio, and video.
• Create a unified semantic layer for entities, metrics, metadata, and relationships.
• Explore self-evolving agents with memory and feedback from executions.

Skills

Python

Java

C++

Golang

LLMs

Agents

Search

Information Retrieval

Education

B.S in Computer Science, AI, ML, or related field

Tools

Elasticsearch/OpenSearch

Milvus

Faiss

Qdrant

Job description

ByteDance in the United States is seeking engineers for the Search team to build Agentic Search and analytics platforms. You will develop multi-source, multimodal search across structured data, documents, images, audio, and video, and unify enterprise data with a semantic layer of entities and relationships.

You will work with LLMs, agents, RAG, and vector databases to enable AI-driven information retrieval at scale.