Principal Software Engineer | AI Platform (Audit Tech) | NYC

Primis is partnered with an AI lab building the platform powering the next generation of audit and accounting products used by professionals on real client work every day.

This role sits within a small, senior team formed from a recent startup acquisition, shipping fast with direct access to product leadership and customers. Think startup ownership with the runway of a public company.

They're hiring a Principal Software Engineer to own the backend and AI orchestration systems that turn frontier models into reliable, production grade workflows at scale.

What you'll work on:

• Own the AI backbone: the backend and orchestration layer powering AI agents, conversations, workspaces, and knowledge base across the platform

• Architect backend services (Python, FastAPI, PostgreSQL, AWS) powering generative AI agents and document centric workflows

• Build and evolve AI orchestration: routing, tool calling, MCP servers, multi step workflows, guardrails, and error handling around third party LLMs

• Design for high throughput, low latency AI workloads: caching, queuing, rate limiting, model failover, and cost/performance tradeoffs

• Work with large scale data: millions of documents, retrieval and search, vector stores, and indexing strategies

• Lead multi quarter initiatives spanning AI, product, and infra, and mentor staff and senior engineers across the org

What we're looking for:

• 8-10+ years of backend development experience building scalable web services and APIs

• Deep Python expertise, with production experience in FastAPI (or similar), PostgreSQL (or equivalent), and a major cloud provider (AWS preferred)

• Strong distributed systems background: data modeling, API contracts, observability, resilience patterns, performance tuning under load

• A track record owning large, complex projects end to end: architecture, execution, rollout, long term operation

• Genuine interest in AI systems and new engineering paradigms: LLMs, agents, retrieval, and what "AI native" software should look like

• Bonus points for hands on LLM API integration (OpenAI, Anthropic), experience with vector databases, embeddings or semantic search, and prior work in regulated domains like fintech, tax, audit, or legal

This role is a particularly strong fit for engineering managers who miss building. You keep the scope and influence of an EM, driving multi team initiatives and setting technical direction, but spend most of your time back in the code and architecture.

Interested? Please apply for more info!

ADVERT DISCLAIMER
Research indicates that men will apply to a role when they only meet 50-60% of the descriptions, however, when looking at women and other minority groups, they can look for up to a 99% match in order to apply to a role. If you feel you are a fit for our role, please still apply, don't worry if you don't tick every single box. We'd still love to hear from you. We encourage underrepresented talent to apply to all our roles and support accessibility needs.