<p>Job Description</p><p><strong>Responsibilities:</strong></p><ul><li>Design, build, and evaluate agentic AI systems that can plan, reason, act, and collaborate across tools and environments, including single- and multi-agent setups for complex, long-horizon tasks.</li><li>Develop robust agent harnesses and evaluation frameworks covering end-to-end testing, regression analysis, trace logging, replayability, and metrics for success, cost, latency, robustness, and safety.</li><li>Implement self-improving agent loops using reflection, critique, self-debugging, and iterative optimization strategies driven by agent experience, execution traces, and automated feedback.</li><li>Architect and optimize agent memory systems, including short-term and long-term memory, retrieval-augmented generation, summarization, compression, forgetting policies, and privacy-aware retention.</li><li>Enable reliable deployment of agents on constrained and edge environments, focusing on model/runtime optimization, partial or offline execution, secure tool-use, and seamless edge-cloud coordination.</li></ul><p>Qualifications</p><p><strong>Basic Qualifications</strong></p><ul><li>Bachelor or master's degree in computer science or engineering</li><li>Strong software engineering skills in Python</li><li>Hands-on experience with LLMs and agentic systems, such as tool-using agents, planner-executor patterns, multistep reasoning pipelines, RAG systems</li><li>Solid understanding of ML fundamentals and practical model usage, including prompting, evaluation, and error analysis</li><li>Ability to design experiments and interpret results, including ablations, statistical thinking, clear success criteria and measurable KPIs</li><li>Strong communication and documentation skills including clear write-ups, reproducible experiments, and crisp technical presentations</li></ul><p><strong>Preferred Qualifications</strong></p><ul><li>3+ years experiences in industrial research.</li><li>Experience with one or more of the following topics:</li><li>Agent frameworks</li><li>Structured generation</li><li>Agent memory management</li><li>Self-improving agents</li><li>LLM fine-tuning & reinforcement learning</li><li>Model optimization for edge</li></ul><p><br></p>
