Senior AI / Agent Engineer

Job Description

Join Bolt as a Senior AI/Agent Engineer to build production AI agents that take real action across our People systems. This is a newly formed team, so you'll be setting the patterns rather than inheriting them. You'll transform complex processes spanning HRIS, ATS, performance, payroll, goals, learning, and data systems into safe, scalable, and reliable workflows. Your work will reduce manual effort and make critical People processes faster, more reliable, and easier to scale.

As the Senior Engineer on the agentic side of the team, you'll own systems end to end 1 from solution design through evaluation, reliability, and production operation 1 and set the patterns the rest of the team builds on. You'll work closely with Integration and Analytics Engineers to define the tool contracts and data requirements your agents depend on, while mentoring engineers and contributing strong engineering patterns. As these systems handle sensitive employee data and employment-related processes, you'll apply a particularly high bar for correctness, auditability, safety, and human oversight.

• Owning the full lifecycle of production AI agents and LLM-powered automation 1 from requirements and solution design through implementation, automated testing and evaluations, release, monitoring, and ongoing maintenance.
• Designing agents that safely take action across People systems, including tool contracts, input validation, scoped permissions, idempotency, retries, rollback strategies, and appropriate human-in-the-loop approval gates.
• Building reliable agentic workflows, multi-step LLM pipelines, and retrieval solutions using event-driven architecture and the orchestration patterns and integration standards adopted by the team.
• Establishing evaluation, observability, and safety practices for your agents, including regression suites, production monitoring, audit trails, cost controls, data minimisation, and safeguards against hallucination or unsafe actions.
• Partnering with People stakeholders, Integration Engineers, and Analytics Engineers to define requirements, protect data integrity, and determine when AI, deterministic automation, integration improvements, or process changes are the right solution.
• Raising the engineering bar through high-quality documentation and code reviews, mentoring engineers, contributing to agentic AI standards and knowledge sharing, and participating in technical hiring.

Requirements

• You have strong software engineering fundamentals, with experience working in service-oriented production codebases and building scalable, fault-tolerant systems.
• You have hands-on experience shipping AI agents or LLM-powered features to production, including agents that take actions across systems using tool/function calling, permission scoping, and safe handling of side effects.
• You're experienced with LLM and agent evaluation, observability, and reliability, and understand how to turn non-deterministic model behaviour into dependable production systems.
• You understand event-driven and asynchronous architectures, APIs, cloud infrastructure, and retrieval approaches such as embeddings, vector stores, retrieval strategies, and re-ranking.
• You have strong technical judgement 1 including knowing when AI is not the right solution 1 and can clearly explain model, agent, and architecture trade-offs to both engineers and non-technical stakeholders.
• You care deeply about engineering quality and responsible handling of sensitive personal data, and you're comfortable mentoring other engineers while designing systems with appropriate security, auditability, transparency, and human oversight.

Company offers

• Accelerate your professional growth with unique career opportunities.
• Enjoy a rewarding salary and stock options, knowing that as Bolt succeeds, so do you.
• Take care of your physical and mental health with our wellness perks.
• Celebrate 5 years at Bolt with a 1-month paid sabbatical to recharge.
• Connect with colleagues at annual company events and smaller team gatherings.
• Balance flexibility and in-person collaboration with our hybrid model, including at least 12 monthly in-office days.