Senior Backend Software Engineer, SandboxAQ

• You are an engineering leader (formally or informally) who thrives on turning nascent ideas into working software that scales
• You are deeply curious about intelligent systems, and how to make them interpretable, auditable, and aligned with human intent
• 5+ years of professional software development experience, ideally with Python as a primary language
• Experience building complex distributed systems, RESTful APIs, and performant backend services
• Proficient in data modeling, query optimization, and have worked with relational and graph databases
• Strong foundation in DevOps, CI/CD, and infrastructure-as-code practices (e.g., Terraform, Docker, GCP preferred)
• Familiarity with agentic AI frameworks (e.g., LangGraph, CrewAI, AutoGen) or orchestration patterns for multi-agent systems
• Strong communication and collaboration skills able to translate abstract requirements into architecture, and architecture into code
• Passionate about building responsibly, you think critically about failure modes, edge cases, and ethical implications of autonomous systems

Desirable

• This is not a typical backend engineering position, but a founding opportunity to define the architecture, culture, and values of a frontier agentic AI platform
• As the first backend engineering hire on our Agentic AI team, you will shape and build the systems that allow intelligent agents to autonomously execute complex scientific workflows, interact with real-world systems, and learn over time
• You’ll be working with a team of world-class scientists and engineers in a high-talent, no-ego environment
• Architect and implement intelligent agents and multi-agent orchestration systems that enable goal-driven automation in scientific research
• Design and build APIs, workflows, and systems that interface with simulations, ML models, third-party tools, and other agents, all while ensuring transparency and safety in their behaviors
• Champion observability, traceability, and reliability making sure every line of code is measurable and maintainable
• Collaborate with AI researchers and domain scientists to prototype agent behaviors in real-world pipelines, and evolve prototypes into robust production systems
• Drive internal engineering standards, infrastructure decisions, and deployment best practices for a research-informed and ethically grounded development process
• Contribute to the broader AI and scientific software community through open-source contributions, publications, or educational initiatives, as time allows