Engineering Manager, Agent Builder

Description

The company is hiring an Engineering Manager to work from its New York office and lead the team building the agent builder for its agentic automation platform for audit and finance.

The agent builder is where firms turn their approach, reference material and know-how into agents, then manage those agents from creation to retirement. A general-purpose agent has no idea how a particular firm audits. What makes an agent work like one of the firm's own staff is that firm's methodology, guidance, document templates and earlier engagement files, and the layer that stores and serves this knowledge is the company's main edge. Nobody has yet worked out how to let auditors build, run and govern production agents in a regulated field, so your team will write that playbook.

The platform runs AI agents in the cloud over hundreds of documents at once, reasoning across PDF, Word and Excel, with a complete audit trail for every step. It has just launched, so much of the work is new, and the team has real room to decide what gets built and how.

The hardest AI problems are less about speed and more about answers people can rely on. In audit, a plausible answer is not enough: every output must be explainable and traceable to its source. The company builds AI that helps audit and finance professionals check evidence more easily so they can spend their time on judgment.

More than 600,000 professionals at 3,000+ organizations in 175 countries use the company's product, including Fortune 500 companies, government agencies and all Big Four audit firms. In 2025 the platform saved customers more than $1.4 billion in productivity. The company raised a $100 million Series B led by Index Ventures at a $1 billion valuation and is a member of 1% for the Planet.

Your team owns the full agent lifecycle, with knowledge at the centre.

Knowledge and context: curating the firm's methodology, guidance, templates, prior-year workpapers and engagement context, with permissions, and serving it to agents at the moment they need it.

Define: the authoring experience and definition model that turn a procedure into a multi-step agent, including steps, tools, knowledge sources, thresholds and human-in-the-loop gates.

Test and publish: evaluating an agent against real cases before release, with review and approval before it reaches the Agent Library.

Run and observe: launching agents on engagements and tracing what they did, which knowledge they used and why.

Govern, version and sunset: permissions, change history and approvals; rolling out new versions without breaking work in progress; deprecating and retiring agents and outdated knowledge cleanly.

Responsibilities

• Set goals by outcome and let the team work out the approach.
• Hire, grow and keep senior engineers in New York, and make the team measurably stronger every quarter.
• Own the technical direction of the agent builder from end to end, staying close enough to the code to make hard calls with credibility.
• Build the knowledge layer that grounds agents in each firm's methodology: how knowledge is captured, kept current, access-controlled and retrieved, and how you prove it improved results.
• Decide how much power builders get and where the guardrails sit, so agents can encode real methodology and still be consistent enough to trust.
• Treat the lifecycle as a product. Versioning, approvals, rollout and retirement are what make firms trust agents at scale, so design them in from day one.
• Build evaluation, observability and auditability into the product rather than adding them later.
• Work with Agent Runtime, Evals, Document Intelligence, Product, Design and audit methodology experts, many based in Amsterdam, to deliver one coherent platform.
• Raise the bar for AI-assisted engineering by taking on harder problems and shaping how engineering uses AI.
• Work with Forward Deployed Engineers, Solution Architects and go-to-market teams, turning what they learn on enterprise rollouts into platform capabilities rather than one-off fixes.

Requirements

• 3+ years leading engineering teams, on top of 5+ years as a software engineer.
• A empowering leadership style and a record of hiring and building high-performing teams.
• Hands-on experience with agentic or LLM systems in production, including context engineering and retrieval. You understand control loops, tool calling, grounding and evaluation well enough to challenge your team's designs.
• Experience building platform products with a lifecycle: versioning, permissions, release and deprecation for things customers build on.
• Strong grounding in software architecture, distributed systems, APIs and cloud-native development.
• Strong product sense: you question framing, make roadmap trade-offs and care whether customers succeed with what you ship.
• Hands-on fluency with AI coding tools and agentic harnesses, and a team that is measurably better for using them.
• Comfortable leading across time zones and offices, with clear written communication.
• Experience in knowledge management or enterprise search: ingestion, chunking, embeddings, hybrid retrieval, document-level permissions and keeping content fresh.
• Knowledge graphs or other structured models of policies, procedures and rules.
• Experience building low-code or configuration products such as workflow designers, rule engines or DSLs.
• Practical agent evaluation: datasets, offline and online experiments, failure classification, and measuring how context affects quality.
• Experience in regulated fields such as audit, financial services or healthcare, where traceability, approvals and access control are required.
• Experience running permissions, approvals and change control across many enterprise customers, teams and thousands of users.
• Python, Azure and observability tooling.
• Domain experience in audit methodology, accounting or fintech.

Conditions

Hybrid work, with at least 3 days onsite at the New York City office.

Hiring process: a 30-minute online recruitment screen, a 30-minute online interview with the hiring manager, a 60-minute online system design interview, a 60-minute on-site product insight discussion, and a 30-minute online final interview.

Hiring is merit-based and open to all, regardless of age, race, ethnicity, nationality, gender, gender identity or expression, sexual orientation, religion, disability or any other characteristic.

Benefits

• Work at one of the fastest-growing, profitable unicorn scale-ups, with global impact.
• Stock appreciation rights to share in the company's success and growth.
• Flexible paid time off.
• Comprehensive medical and dental coverage.
• 401K match.
• Paid parental leave.
• Company-sponsored lunch.
• Learning and development programmes to build your skills.
• A global team spread across five offices.
• Access to a mental health and wellness platform.
• Team social activities, including company-wide events such as the GO gathering, where global teams come together.