Baker Jobs

About The Position

Requirements

• Bachelor’s degree in computer science, engineering, or a related field, or equivalent practical experience.
• 15+ years of software engineering experience, including 3+ years building and deploying LLM-based or agentic systems in production.
• Staff- or principal-level engineering experience, with demonstrated success personally designing and shipping production agent systems.
• Deep experience designing and building evaluation harnesses and using evaluation results, baselines, and regression testing to drive iteration and release decisions.
• Strong knowledge of modern agentic technologies, including orchestration frameworks, tool/function calling and MCP, vector- and graph-based retrieval, and context and memory management.
• Experience with evaluation methodologies, including evaluation design, baselines, regression discipline, confidence frameworks, and LLM-as-judge patterns and limitations.
• Full-stack product engineering experience, including building applications and user experiences in which agents operate.
• Experience designing human-in-the-loop workflows, checkpoints, review experiences, and failure-containment mechanisms.
• Strong hands-on software engineering skills with recent experience writing production-quality code in Python and/or TypeScript.
• Experience with observability and tracing for agentic systems.
• Experience directing third-party or partner engineers against established technical standards.
• Experience evaluating third-party technologies or vendor agents against internal technical and quality standards.
• Demonstrated ability to use AI and agentic development tools to multiply effectiveness in day-to-day engineering work.
• Strong ability to collaborate across engineering, product, design, platform, and business teams while translating emerging technical practices into repeatable standards.
• No certifications or licenses required.

Nice To Haves

• Advanced degree preferred.

Responsibilities

• Define and build agentic reference patterns for task decomposition, orchestration, context engineering, memory management, tool use, human-in-the-loop checkpoints, and failure containment and recovery.
• Maintain the agent pattern library and establish orchestration standards for product teams.
• Own BT Compass end to end, including its agentic workflows and full-stack product experiences, on the intelligence layer built by the Data & Intelligence team.
• Build and own the evaluation harness, baselines, and regression discipline used to drive iteration and inform release and acceptance decisions for internally developed and third-party agents.
• Establish evaluation and acceptance criteria in partnership with Product, Architecture, Platform & Production Engineering, Quality Engineering, and other stakeholders.
• Partner with Platform & Production Engineering to enable evaluation capabilities to operate at scale in production.
• Lead the growing Agent Factory and set standards for agent design, engineering quality, and full-stack product development.
• Conduct hands-on design and code reviews across product teams and mentor engineers in agentic engineering practices.
• Hire and develop full-stack, forward-deployed AI engineers who build both agents and the applications in which they operate.
• Partner with expert agentic-engineering resources and translate leading practices into Baker Tilly-owned standards, patterns, and runbooks.
• Direct third-party and partner engineers delivering solutions against Baker Tilly’s technical standards.
• Evaluate third-party and vendor agents against the firm’s technical and evaluation standards.
• Partner with the Chief Architect to establish evaluation standards and promote proven patterns into the platform’s shared library.
• Collaborate with Data & AI Platform, Product, Design, Quality Engineering, Platform & Production Engineering, subject matter experts, and engineering teams in India.
• Engage in special projects and perform other duties as assigned.

Benefits

• Compensation and benefits package for eligible employees.