Senior Staff AI Engineer, Agentic AILocation: Bay Area - Oakland, CA
Employment Type: Full-time
Experience: 8-15 years
Focus: Agentic AI, Scientific Reasoning, AI Harnesses, Tool Use, Memory, Evaluation Systems
About Our ClientOur client is building the first AI-native operating system for materials science.
Their platform helps scientists, chemists, and R&D teams reason through complex scientific problems using AI-native workflows, proprietary scientific data, and agentic systems designed for real-world research environments.
The company is working with a uniquely valuable scientific data foundation, including trillions of proprietary scientific tokens that are not available anywhere else. This creates a rare opportunity to build AI systems that understand how scientists think, work, and make decisions.
About the RoleOur client is hiring a
Senior Staff AI Engineer to own the technical direction of the agentic AI harness at the center of the platform.
This is the most senior hands-on individual contributor role on the Agentic AI team. This person will define how the product evolves from a chat-based experience into an agentic-first platform where AI agents can reason, plan, use tools, remember context, evaluate outcomes, and operate across the full scientific workflow.
This is a true 0-to-1 role. You'll build the orchestration, tool use, memory, planning, and reasoning layer for production agent systems while helping lead two junior engineers on the AI harness. You'll also partner closely with product and ML leadership to expand agent capabilities across every surface of the platform.
What You'll Do- Own the technical direction for the agentic AI harness
- Build and expand production agent systems for scientific and chemistry workflows
- Design orchestration, tool use, memory, planning, and evaluation layers for agentic systems
- Help shift the platform from a chat-style interface into an agentic-first experience
- Build tools and sub-agent personas that reflect how chemists think and work
- Integrate frontier reasoning models into the application
- Expand agent capabilities across scientific workflows and product surfaces
- Contribute to potential fine-tuning programs for domain-specific scientific reasoning
- Make agent behavior observable, measurable, and reliable through tracing and evaluation systems
- Evaluate agent performance against real scientific tasks
- Drive agent reliability to the level where customers can trust agents to act with increasing autonomy
- Provide technical leadership to junior engineers on the AI harness team
- Partner closely with product, ML, and engineering leadership on roadmap and architecture
What We're Looking For- 8-15 years of experience in AI engineering, software engineering, ML engineering, or related technical roles
- Staff-level or Senior Staff-level experience setting technical direction, not just executing tasks
- Experience building sophisticated production agent systems from inception through scale
- Experience building an agent harness or similar agentic infrastructure
- Hands-on experience with low-level agentic frameworks such as LangGraph, LangChain, or equivalent tools
- Strong full-stack programming experience across Python, React, TypeScript, or similar technologies
- Deep understanding of LLMs, agent orchestration, tool use, memory, planning, evaluation, and production reliability
- Ability to design systems that move beyond chat into autonomous or semi-autonomous workflows
- Strong judgment around system architecture, model behavior, observability, and product safety
- Experience working in a startup, or a background combining big tech experience with startup execution
- Ability to lead technically while remaining deeply hands-on
Technical EnvironmentRelevant technologies and concepts include:
- Python
- LangGraph
- LangChain
- Agentic frameworks
- Frontier LLMs, including GPT-4, Claude, and similar models
- Braintrust
- Tracing and evaluation systems
- RAG
- React
- TypeScript
- Tool use
- Memory systems
- Agent orchestration
- Scientific reasoning systems
Preferred BackgroundOur client is especially interested in candidates with:
- A degree in science, engineering, computer science, or a related technical field
- Master's or PhD from a strong technical university
- Scientific background or professional exposure to chemistry, physics, biology, materials science, or mechanical engineering
- Experience fine-tuning reasoning models
- Experience building AI systems for scientific reasoning
- Experience working with proprietary datasets or domain-specific AI systems
- Experience mentoring or leading junior engineers while staying hands-on
Why This Opportunity- Own the agentic AI layer at the center of an AI-native scientific platform
- Build agent systems for real chemistry and materials science workflows
- Work with proprietary scientific data that no one else has access to
- Help define how scientists and AI reason together
- Move a platform from chat-based AI into agentic-first workflows
- Partner directly with product and ML leadership on company-defining technical direction
- Step into a Senior Staff-level role with broad technical ownership and meaningful product influence
- Build at the intersection of frontier LLMs, agentic systems, scientific reasoning, and enterprise R&D
Ideal Candidate ProfileThe ideal candidate is a Senior Staff-level AI engineer who has already built production agent systems and knows what it takes to move them beyond version one.
They are deeply hands-on, technically opinionated, and capable of setting direction across agent orchestration, memory, tool use, evaluation, tracing, and product integration. They understand that reliable agent behavior requires more than model calls - it requires systems, feedback loops, observability, and strong product judgment.
This person should be excited by the opportunity to build AI systems that help scientists reason, discover, and work in fundamentally new ways.
Never miss an opportunity! Create an alert based on the job you applied for.
