About this role
Design and develop core Agent logic and systems—including task planning, tool calling, multi-turn dialogue, memory, RAG, context engineering, and multi-agent collaboration—and lead model training, evaluation, and deployment for multimodal, personalized AI applications.
Skills for this role
Task planningOrchestrationTool callingMulti-turn dialogue managementMemory architecturesRAGContext engineeringMulti-agent collaborationContinuous pre-trainingPost-trainingDataset creationData pipelinesReinforcement learningRL reward modelingInstruction followingReasoningPrompt engineeringMultimodal conversationLong-term memoryShort-term memoryLong-context handlingRecallRankingLong-text processingMulti-document synthesisAPI integrationMCP integrationContext managementQuantitative evaluation designPyTorch