About FirmPilot
FirmPilot builds AI-powered systems that automate and scale real-world business outcomes. We focus on applied AI, using best-in-class large language models and tooling to deliver reliable, production-grade automation.
We are not building bespoke foundation models. We are building systems that use AI effectively, which requires both AI expertise and strong software engineering.
The Role
We are hiring a Software Engineer with applied AI expertise to design and optimize the integration of LLMs into production systems.
This role is for engineers who understand how LLMs behave in practice and can build repeatable, testable, API-driven workflows around them. This is not a role for casual AI experimentation or “prompt tinkering.”
What You’ll Do
- Design, implement, and evolve RAG pipelines combining structured data, embeddings, and LLMs.
- Develop and maintain prompt strategies used across multi-step agent workflows.
- Integrate LLMs into production systems with attention to reliability, cost, and latency.
- Orchestrate AI-driven workflows across internal services and third-party APIs.
- Define and improve practices for prompt versioning, evaluation, and iteration.
- Work closely with platform engineers to ensure AI systems are maintainable and observable.
- Help the team reason about when deeper customization or model-level work may be warranted in the future.
What We’re Looking For
- 5+ years of professional software engineering experience.
- Strong backend experience with modern .NET and Python.
- Demonstrated, hands-on experience with:
- Retrieval-Augmented Generation (RAG)
- Prompt engineering beyond simple completion calls
- Vector databases and embedding workflows
- API-driven LLM integrations
- Experience integrating AI systems into real products, not just prototypes.
- Ability to reason about trade-offs: determinism, accuracy, cost, and performance.
- Strong communication skills and the ability to explain AI behavior clearly.
Our Engineering Environment
- Modern .NET (8 & 9), Python, TypeScript
- GraphQL and REST APIs
- LLMs, embeddings, and vector databases
- Event-driven, agent-oriented architecture
- AWS as our primary cloud provider
- Terraform for infrastructure as code
- CI/CD with automated testing and observability
How We Work
- AI systems are treated as production software, not experiments.
- We move quickly by setting clear standards and iterating frequently.
- Communication is frequent and expected, especially when behavior is unclear.
- Engineers are expected to challenge assumptions and improve systems collaboratively.
Not a Good Fit If…
- You are looking for pure AI research or model training work.
- You primarily use AI tools without understanding how they work internally.
- You are uncomfortable debugging non-deterministic systems.
- You prefer slow iteration or minimal ownership.
Pursuant to applicable pay transparency requirements, the Company discloses in good faith that the anticipated salary range for this role is $135,000 to $162,000.
