Senior Applied AI Engineer

Setting up a 401(k) shouldn’t be the hardest thing a business ever does. But retirement plan administration is complex, fragmented, and manual — and it slows down both business owners and the advisors who serve them. Aboon was founded to change that.

We’re an AI-native third-party administrator (TPA) that makes it easy for advisors to offer 401(k) s and other retirement products by pairing streamlined workflows with dedicated human service. Backed by leading investors including Bain Capital and Edward Jones Ventures, we bring decades of experience building for advisors and the retirement industry. We’re looking for exceptional, mission-driven people to build it with us.

Most retirement plan work begins with messy, fragmented inputs gathered from multiple sources, including employee census files, prior plan documents, payroll files, recordkeeper forms, distribution forms, and emails between advisors and business owners. Plan administrators have to review, parse, and validate these materials, then determine what actions to take. The work is time-consuming, and mistakes can be costly.

As a Senior Applied AI Engineer at Aboon, you’ll build and operate the agent systems that handle this work. You’ll develop systems to extract and validate data from PDFs and spreadsheets, connect agents to recordkeeper, payroll, and internal systems, and keep workflows reliable when inputs are incomplete or steps fail. You’ll own your systems from design through evaluation, deployment, and ongoing improvement in production.

You’ll shape how we build with AI through the systems you ship and the standards you set.

You’ll be a hands-on individual contributor, reporting to our Head of Engineering and working closely with engineers and retirement plan experts who understand the rules and edge cases.

• Turn messy inputs into reliable data. Extract and validate structured data from census files, plan documents, PDFs, and spreadsheets to support plan design, onboarding, and compliance.

• Build agents that complete workflows. Design systems that use tools and external services, track state, recover from failures, and hand off to a person when needed.

• Measure quality. Build evaluation datasets, regression tests, and production monitoring so you can track quality over time and catch problems before users do.

• Decide where AI belongs. Choose between a model, deterministic code, retrieval, or human review based on what each step needs, and explain the tradeoffs.

• Make production systems cheaper, faster, and more reliable. Reduce failures, latency, and cost, and make systems easier for the team to understand and debug.

• Raise the bar for the team. Write design docs, review code, and help other engineers build and test high-quality AI features.

• You work well with non-technical experts. Much of this job is sitting with plan administrators, learning how they make decisions, and turning that into workflows and agents.

• You write clearly. Design docs, code reviews, and explaining tradeoffs to people outside engineering are a real part of the role.

Required

• 6+ years building and running production software, with clear ownership of technical decisions and what happened after launch.

• You’ve shipped an LLM- or agent-powered system that people used in a real workflow. You can explain what you built, how you measured quality, and what you changed when it failed.

• Experience testing system behavior with tests, evaluation datasets, production monitoring or structured human review.

• You reason about state, retries, fallbacks, permissions, latency and cost, not just model outputs.

• You’re comfortable integrating with external systems and working with both structured and unstructured data.

• You’ve…