Staff AI Engineer

This is a fully remote position, open to applicants in California.

📋 Description

• Lead the development of agentic AI, focusing on multi-agent orchestration patterns, agent-to-agent communication protocols, and dependable tool utilization at a production level.

• Take charge of prompt engineering and evaluation processes, which include structured outputs, minimizing hallucinations, and ensuring behavioral consistency.

• Construct and manage MCP server infrastructure that provides backup data to AI agents through the Model Context Protocol, ensuring tenant-aware RBAC, query limitations, and secure tool boundaries.

• Establish AI quality benchmarks for retrieval relevance, summarization accuracy, and agent dependability, while promoting systematic enhancements through evaluation-driven iterations.

• Advocate for security and safety in AI systems, encompassing adversarial prompt fortification, jailbreak resistance, data boundary enforcement, and awareness of the OWASP LLM Top 10.

• Optimize AI workflows for performance, cost-effectiveness, latency, and observability across billions of documents in various global locations.

• Mentor team engineers, elevate the technical standards, and participate in architecture assessments and design decisions.

⛳️ Requirements

• Demonstrated experience in integrating AI/ML services and APIs into production backend systems, including APIs, asynchronous pipelines, and cloud infrastructure, treating models and inference endpoints as integral components of a broader service architecture.

• Practical experience in deploying LLM-powered functionalities from start to finish, such as embeddings pipelines, RAG, summarization, or LLM-powered assistants, with a solid grasp of potential failure modes.

• Proven background in designing and managing multi-step agentic workflows that utilize real tools, including strategies for reliability, observability, and recovery.

• Familiarity with the Model Context Protocol (MCP), including the construction of MCP servers, devising tool exposure contracts, or integrating MCP into agent workflows.

• Hands-on experience with prompt engineering and evaluations, involving structured outputs, hallucination reduction, evaluation frameworks, and LLM observability tools.

• Experience in developing multi-tenant systems with data boundary enforcement, tenant-aware access controls, and safety measures for LLMs.

• Understanding of authentication and authorization patterns, including OAuth 2.0, OIDC, JWT, and API key management in cloud backend systems.

🏝️ Benefits

• Unlimited paid time off, 12 paid holidays including 4 global VeeaMe Days dedicated to self-care, and 24 paid volunteer hours annually through Veeam Cares.

• Paid parental leave: 8 weeks for all parents and 16 weeks for birthing parents.

• Comprehensive medical, dental, and vision coverage commencing on your first day.

• Mental health support, therapy sessions, and digital wellness tools available through our Employee Assistance Program.

• 401(k) retirement plan with company matching contributions.

• Support for fertility, adoption, and surrogacy through Maven, along with paid volunteer time.

• AirVet: Access to 24/7 virtual veterinary care at no cost.

• Legal services, identity protection, and additional health insurance options.

• Tax-advantaged spending accounts for healthcare, dependent care, and commuting expenses.

• Opportunities for learning and development through on-demand libraries (LinkedIn Learning, O’Reilly), mentoring, workshops, and learning events such as our annual Global Day of Learning.