About this role
Design and research architectures that continuously evaluate and enhance their own performance by building feedback-driven systems which detect weaknesses, generate corrective hypotheses, and implement improvements via reflection, retraining, or workflow adaptation. Study self-modeling, reflective reasoning, reward modeling, and techniques bridging reinforcement learning, interpretability, and meta-optimization to enable continuously learning enterprise systems.
Skills for this role
Feedback-driven systemsEvaluatorsRetrieversIterative refinementPerformance dashboardsReflectionRetrainingWorkflow adaptationSelf-modelingReflective reasoningReward modelingSelf-evaluationReinforcement LearningInterpretabilityMeta-optimizationCompound AI systemsAgentic collaborationEnsemblingReActGraph-of-thoughtsChatGPTCursorPerplexityProgrammingData analysisPrototype developmentExperimental evaluation