Senior AI ML Developer with Agent Memory

DataArt·Armenia, Bulgaria, Cyprus, Georgia, Kazakhstan, Latvia, Poland, Romania, Serbia, Ukraine·Удалённо, Офис·сегодня

About the Position

We are looking for a Senior AI ML Developer to help design and implement memory and context management capabilities for conversational AI systems. In this role, you will work on agent memory strategies, multi session personalization, and observability for memory effectiveness and growth.

About the Project

You will contribute to a Conversational AI Platform centered on memory enabled agent design and context management. The work includes improving how agents store, retrieve, and manage summaries, user facts, and preferences while supporting retention and compliance needs.

Responsibilities

  • Design memory and context management for LLM agents or chatbots
  • Implement short term and long term agent memory strategies
  • Configure memory enabled agent design without custom infrastructure
  • Develop memory extraction approaches for conversation summaries and user facts or preferences
  • Manage agent memory lifecycle including write on turn, retrieve on turn, relevance filtering, and TTL retention
  • Build multi session state management and personalization capabilities
  • Track memory writes, retrieval effectiveness, and storage growth
  • Support data retention policies and GDPR user deletion requirements

Requirements

  • 5 plus years building AI ML systems or conversational AI platforms
  • Experience designing memory or context management for LLM agents or chatbots
  • Experience with multi session state management and personalization systems
  • Knowledge of data retention policies and compliance including GDPR user deletion
  • Understanding of agentic memory taxonomy including short term and long term memory
  • Knowledge of memory extraction strategies including summary and semantic approaches
  • Experience with agent memory lifecycle management

Nice to Have

  • Hands on experience with AWS AgentCore Memory
  • Experience with LangChain or LangGraph memory modules
  • Experience with vector databases for semantic memory including Pinecone, Chroma, or FAISS
  • Knowledge of memory consolidation and relevance scoring

Technologies

AWS AgentCore Memory, LangChain, LangGraph, Pinecone, Chroma, FAISS

Похожие вакансии

Другие вакансии DataArt