Staff Software Engineer, AI

This is a fully remote position, open to applicants in United States, +1 more country.

📋 Description

• Establish the technical vision for Compass and its LLM-driven features.

• Design multi-agent systems and oversee significant design choices.

• Lead the architecture and development of MCP servers, A2A integrations, and APIs that connect agents to the Basis media buying toolkit.

• Manage agent orchestration, evaluation, and observability.

• Define quality metrics for agents and ensure dependable production reasoning, planning, and execution.

• Take full responsibility for software from start to finish, including production support for Compass.

• Act as the main technical liaison for Product, QA, Design, Operations, Engineering teams, and stakeholders.

• Represent the team in cross-departmental planning, architectural discussions, and RFCs.

• Lead extensive technical projects across various teams and systems from conception to production.

• Manage team deliverables, decisions, quality, processes, and technical strategy.

• Convert business objectives into actionable technical plans while identifying risks and obstacles.

• Uphold the technical quality standard through thorough design and code reviews.

• Facilitate the onboarding and mentoring of engineers.

• Champion the adoption of AI-assisted and agent-based development within Basis Engineering.

• Establish best practices and create examples and infrastructure for other teams.

• Cultivate a positive team environment and promote Basis within the community.

• Develop orchestration, tooling, and evaluation infrastructure for dependable LLM-powered agents in the advertising technology sector.

• Create patterns and best practices for agentic and AI-assisted development throughout Engineering.

⛳️ Requirements

• At least 8 years of professional software engineering experience, with recent involvement in delivering production-level LLM-powered applications or agent systems.

• Proven record of technical leadership, including directing the technical trajectory of a team or major system and impacting architecture beyond immediate responsibilities.

• Expert-level knowledge of Python.

• Experience in designing production APIs and backend services, as well as integrating with protocols such as MCP and A2A.

• Extensive background in building and managing production-quality agentic systems using frameworks like Strands, LangChain, or the Claude Agent SDK.

• Familiarity with asynchronous and event-driven models, prompt engineering, structured outputs, and methods to enhance agent reliability.

• Experience with evaluation and observability tools like Langfuse or LangSmith, including creating evaluation strategies from scratch.

• Proficiency with AI coding tools such as Claude Code, Cursor, and Codex.

• Demonstrated ability to mentor engineers and clearly articulate complex agentic systems to both technical and non-technical stakeholders.

• A collaborative, design-conscious, and decision-making approach that emphasizes equity, access, and continuous learning.

• Dedication to fostering inclusive and respectful environments.

• A Bachelor’s or advanced degree in Computer Science, Engineering, or a related field is a plus, but not mandatory.

🏝️ Benefits

• Flexible work week.

• 401k/RRSP matching.

• Mental health support.

• Paid sabbaticals.

• Generous parental leave.

• Flexible work options.

• Competitive benefits package.

• Options for remote, hybrid, or on-site work throughout the U.S. and Canada.

• Optional cover letter.

• Verification of identity, education, and employment history for new hires.