Senior Software Engineer - Agentic AI Systems

Cognichip is seeking a Senior Software Engineer to design, implement, and deploy advanced agentic AI systems. In this high-impact role, you will build production-ready AI agents and partner with research and engineering teams to translate complex chip design workflows into reliable, scalable solutions. Responsibilities: Build Agentic Systems – Implement multi-step reasoning agents with advanced memory, Retrieval-Augmented Generation (RAG), and integrations to tools, databases, and APIs; Evaluate and Optimize Agent Performance – Define and implement evaluation pipelines for agentic systems, including success/failure classification, grounding accuracy, reasoning robustness, tool-use reliability, and long-horizon task completion. Use metrics and benchmarks to continuously improve performance in production environments; Orchestrate & Optimize – Design supervisor/sub-agent patterns, enable coordination across agents, and apply best practices for robustness, performance, and scalability; Deploy & Evolve – Deliver production-grade agentic AI workflows on cloud platforms (AWS Qualifications: Bachelor's or Master's degree in Computer Science, Software Engineering or a related field; 5-10 years of experience in software development in cloud environment; 2+ years of experience with hands-on experience building and deploying production-grade agentic AI systems with real-world applications; Solid working knowledge of Retrieval-Augmented Generation (RAG), agent performance evaluation, and hands-on experience with LangGraph and LangSmith platforms; Proficiency in Python and familiarity with backend cloud services (preferably AWS); Working in an onsite setting; Authorized to work in the United States Required Skills: Agentic AI Systems, Multi-agent orchestration, Supervisor patterns, LangChain, LangGraph, LangSmith, Retrieval-Augmented Generation (RAG), Agent performance evaluation, Cloud environments, AWS, Containerized deployment, Backend integration, Python, Backend cloud services, Software engineering practices, GitHub, CI/CD pipelines, Testing frameworks