Platform Engineer - AI

• About the role
• HWA International builds the trust accounting software that bank trust departments, credit unions,
• foundations, and family offices run their fiduciary operations on. We are an AI-first engineering
• organization, and agentic coding tools are how we build. Our platform is in a significant modernization,
• and this role is one of the engineers who builds it.
• Agentic tooling lets us understand existing behavior and specify what comes next far faster than that
• work could be done by hand. You will use it to do exactly that, then build the modern platform end to
• end: the C# and .NET services, the web front end, and the AWS infrastructure they run on. This is a full-
• stack role, and you own the infrastructure your work runs on.
• Accounting accuracy is the product, so correctness is the bar. Anything new has to produce the same
• results as what it replaces, or better, and proving that is part of the job.
• You will work within an architecture set by engineering leadership, with real influence on it. We work to
• modern engineering standards: trunk-based development on GitHub, continuous integration,
• infrastructure as code, and feature flags.
• What you will do

• Build the modern platform across the stack. Write features, services, and fixes in C# and .NET,

and the web front end that sits on them, against the architecture and patterns the team has set.

• Own the infrastructure for what you deliver. Within the cloud architecture engineering

• leadership sets, define and maintain your AWS environments as code, along with the
• networking, managed databases, and services behind them, so every environment is
• reproducible and version controlled.

• Own delivery and observability. Build and run the CI/CD pipelines, staged rollout, and feature

flags that take your work to production, and the monitoring, logging, and alerting that tell us

how it behaves there.

• Work out what existing code does and write it down. Use agentic tooling to analyze established

application and database logic, then produce a clear written specification of the behavior that

has to be preserved.

• Move business logic into application code and prove it matches. Relocate rules wherever they

• live today, including inside the database, into maintainable, tested application code. Build the
• data migration paths and test harnesses that show the modern platform's calculations match
• the existing ones.

• Verify what the tools give you. Agentic tooling drafts the analysis, the documentation, and

much of the code. An agent's account of what existing code does is a hypothesis until you have

read the code and confirmed it. Characterization tests come before behavior changes.

• Own the work in production. Whatever ships under your name is yours, from the first analysis

• through to how it behaves in production. Take part in incident response, communicate clearly
• while an issue is open, and follow through to root cause and a durable fix.
• What we are looking for
• You do not need to be an expert in every layer. Agentic tools let one strong engineer work across the full
• stack and the infrastructure under it, so we look for depth where judgment matters most, enough
• working knowledge everywhere else to direct the tools and catch their mistakes, and fluency with the
• tools from day one.
• Where you need depth

• Several years of professional C# and .NET development, with production code you can describe

and defend.

• Strong relational database skills: schema design, query tuning, and correct transaction and error

handling by reflex. PostgreSQL experience strongly preferred.

• Demonstrated ability to work out what unfamiliar code does, including code written in a

language or platform you have not used before, and a repeatable method for doing it safely.

• Agentic coding is table stakes. Deep, hands-on fluency with agentic tools (Claude Code, GitHub

• Copilot, Cursor, or similar) as your primary way of working on real production code, from day
• one, with the judgment to treat what they produce as work to be checked rather than answers
• to be shipped.
• Where working knowledge is enough

• Front end: a modern web framework, React with TypeScript preferred.
• Cloud: AWS (Azure considered), infrastructure as code, networking, managed databases, and

containers.

• Delivery: CI/CD with automated testing and zero-downtime deployment, plus production

monitoring and alerting.

Across everything

• Git and GitHub, with trunk-based development and short-lived branches.
• Comfort owning quality on a system that cannot fail quietly: disciplined testing, observability,

and careful, staged rollout.

• Clear written and spoken English, because much of your working contact with the team is

written.

• Availability to work hours that overlap the first half of the US Eastern working day, so you work

with your counterparts rather than around them.

Nice to have

• Experience in financial services or another regulated, high-stakes domain. Trust accounting,

wealth management, or fiduciary services is a strong plus but is not required. We will teach the

domain to the right engineer.

• Experience modernizing an established business application onto a cloud-native stack.
• AWS certification or equivalent depth in security, identity, and cost management on AWS.
• Building software under SOC 1 / SOC 2 or comparable regulated-environment controls.