Software Engineer - Agentic AI Systems

Job Title

Staff Software Engineer - Agentic AI Systems

• We are seeking a Staff Agentic AI Engineer to lead the architecture, implementation, and production deployment of advanced agentic AI systems.
• In this role, you will serve as a technical authority for multi‑agent systems across Cognichip, driving long‑horizon autonomous workflows that integrate proprietary models, semiconductor design tools, and cloud infrastructure.
• You will design systems that reason across multiple steps, manage memory and knowledge grounding, and operate reliably in production over extended periods.
• This is a senior individual contributor leadership role.
• You will define architectural patterns, raise engineering standards, mentor other engineers, and partner closely with Applied AI, Product Engineering, and Platform teams to translate cutting‑edge research into scalable enterprise solutions.
• Success in this role is measured not by prototypes, but by robust, production‑grade agentic systems shipped to customers.

• Own the end‑to‑end architecture of agentic AI workflows, including reasoning pipelines, memory systems, RAG, evaluation frameworks, and orchestration patterns.
• Define best practices for supervisor/sub‑agent coordination, fault tolerance, long‑horizon reasoning, and system robustness.
• Serve as Cognichip’s internal expert on agentic AI system design and production deployment.

• Design and implement multi‑step autonomous agents with advanced memory, Retrieval‑Augmented Generation (RAG), and integrations to tools, APIs, and enterprise data sources.
• Deliver production‑grade workflows deployed on cloud platforms (AWS preferred), with strong observability, monitoring, and reliability guarantees.
• Drive continuous improvement of agent quality, cost efficiency, and performance in real customer environments.

• Define and implement comprehensive evaluation pipelines for agentic systems:
• Task success / failure classification
• Grounding accuracy
• Reasoning robustness
• Tool‑use reliability
• Long‑horizon completion rates
• Establish regression testing and benchmarking strategies using frameworks such as Lang Smith or custom evaluation infrastructure.
• Balance automated evaluation with human‑in‑the‑loop feedback for complex workflows.

• Partner with Applied AI researchers to product ionize new capabilities.
• Work with backend/platform engineers to integrate agents with cloud infrastructure and enterprise systems.
• Collaborate with product managers to translate semiconductor workflows into agent‑driven user experiences.

• Set technical direction for agentic AI systems across teams.
• Mentor senior and mid‑level engineers.
• Raise engineering standards around agent architecture, evaluation, and production readiness.

• Bachelor’s or Master’s degree in Computer Science, Software Engineering, or related field.
• 8–12+ years of professional software engineering experience.
• 3+ years building and deploying production‑grade agentic AI systems.
• Deep hands‑on experience with:
• Multi‑agent orchestration frameworks (Lang Graph, Lang Chain, Lang Smith, or equivalents)
• RAG pipelines and memory systems
• Agent evaluation methodologies
• Strong proficiency in Python and backend cloud services (AWS preferred).
• Proven track record delivering complex AI systems into production.

• Contributions to open‑source AI projects or frameworks.
• Experience with multi‑agent orchestration patterns at scale.
• Knowledge of reinforcement learning, planning algorithms, or autonomous reasoning.
• Track record of deploying agentic AI systems in production at scale.

• The chance to work on state‑of‑the‑art AI systems that push the boundaries of autonomy and reasoning.
• A collaborative environment where engineering meets research.
• Competitive compensation and equity in a fast‑growing AI startup.
• A culture that values ownership, curiosity, and technical excellence.