Principal Software Development Engineer - AI
Summary
Principal-level engineer leading architecture and hands-on build of production LLM/agentic systems across a full product stack (Python back end, modern front end, .NET a plus). Day to day: designing distributed microservices, defining AI platform boundaries (MCP/A2A orchestration, eval-gated CI/CD, pgvector persistence) and mentoring senior engineers in a regulated money-movement domain.
Title: Principal Software Development Engineer - AI
Location: Remote - able to start 9am EST time
Duration: Permanent - Direct hire
Requirements:
- Professional software engineering experience with progressively increasing impact, including multiple years designing and operating production LLM/agentic systems at Staff-level scope or above.
- Proven technical leadership designing distributed systems and microservices for complex domains, with working knowledge of DDD and Clean Architecture principles applied to build highly performant, well-architected systems (prior DDD project experience not required).
- Full-stack breadth, spanning modern front-end frameworks and a Python back-end, sufficient to lead architecture across the entire product stack, not just AI services. .NET/C# experience is a plus, not a requirement.
- Track record of authoring architecture standards adopted beyond your own team, producing decision records that other senior engineers review, cite, and build on.
- Demonstrated quantified build-vs-buy decisions against named vendors (e.g., eval/observability platforms, vector stores), with cost models and explicitly rejected alternatives.
- Deep fluency in the current AI platform landscape: stateful agent orchestration, MCP and inter-agent (A2A) interoperability, eval-gated CI/CD, and vector-enabled persistence (PostgreSQL/pgvector, DiskANN-class indexing, managed offerings).
- A considered point of view on where the companies model strategy should be in two years, including whether and where to adopt fine-tuned small language models versus frontier APIs, grounded in cost, latency, and control tradeoffs.
- Experience defining platform boundaries between AI stacks and an existing product stack (e.g., Python AI services alongside .NET), including shared service-kit libraries other teams consume.
- Security and governance leadership: resource isolation, agentic workflow guardrails, responsible AI in a regulated, money-movement domain.
- Executive communication, articulating platform tradeoffs in terms leadership can act on, covering cost, risk, and optionality, not just engineering detail.
- A high degree of agency, building net-new systems and optimizing API performance where no established pattern exists, forging the path rather than waiting for one, with demonstrated ability to bring other engineers along that path.
- Demonstrated multiplier effect through mentorship: pairing with, unblocking, and growing senior engineers.
- A passionate and key contributor to the companies software factory, directing and reviewing agentic development while maintaining and improving our high quality and compliance bar.