Research Scientist/Engineer - Agent Memory

The role

This person will design and develop Agent Memory products and technical architectures to store and utilize work context for individuals and organizations. They will extract information from conversations and documents and develop methods for memory updates, merging, and retrieval.

What you will do

• Design and develop Agent Memory products and technical architectures to store and utilize work context.
• Extract and structure information to be remembered from conversations, documents, and work history.
• Develop methods for updating, merging, deleting, and managing the freshness of existing memory.
• Develop functions to search for memory suitable for user requests and situations and provide it as Agent Context.
• Design a common Memory Layer where multiple AI Agents can share and inherit individual and organizational knowledge and work context.
• Verify the latest Agent Memory research and open source and implement them into products for actual users and organizations.

Required qualifications

• 4 or more years of research or development experience in NLP, LLM, or software or equivalent capabilities.
• Bachelor's degree or higher in Computer Science, Artificial Intelligence, Software, or a related field or equivalent practical skills.
• Experience proactively developing products or systems using one or more of LLM, NLP, RAG, search, Knowledge Graph, or personalization technologies.
• Experience independently performing one or more technical tasks from design to implementation, evaluation, and product application.
• Ability to develop product or service-level software using Python.
• Experience designing and developing REST API or asynchronous Backend systems.
• Ability to define requirements into specific MVPs and step-by-step development scopes.
• Ability to define experimental hypotheses and evaluation criteria and analyze quantitative and qualitative results.
• Ability to understand papers, open source, and technical documents and apply them to actual systems.
• Experience with Git-based collaborative development and code review.

Nice to have

• Master's or Doctoral degree in Artificial Intelligence, Natural Language Processing, or related fields.
• Experience implementing Agent Memory or long-term conversation Context management functions.
• Experience developing user profiles, personalization, recommendation, or long-term user modeling systems.
• Experience using search and knowledge structuring technologies such as Keyword Search, Vector Search, Hybrid Search, Re-ranking, or Knowledge Graph.
• Experience designing memory deduplication, conflict resolution, update, forgetting, or deletion policies.
• Experience building evaluation datasets and automatic evaluation systems for LLM or search systems.
• Experience operating AI services in FastAPI, gRPC, Docker, Kubernetes, or Cloud environments.
• Experience in open source contribution, paper publication, patent application, or technical document writing.

Before you apply

• Apply through https://www.lgresearch.ai/careers.
• Technical interview includes a portfolio presentation focused on personal Research area.
• LG AI Research is not a military service designated company, so recruitment and transfer of specialized research personnel is not possible.

About the company

Language Lab researches core advanced technologies to solve technical challenges in speech and language processing. The Agent Architecture Tech Cell under Language Lab develops Agent Harness and Agent Memory to enable stable AI Agent execution and continuous use of work context.