Applied AI Engineer

ABOUT the company

The company builds AI software for banks: purpose-built small language models, a banking ontology, and AI bankers that financial institutions can trust. Our models outperform general-purpose LLMs by 30 to 80 percent on banking tasks. We operate under the compliance, audit, and model-risk standards that banking requires.

WHY THIS ROLE EXISTS

The company is growing from a handful of live banking customers to thirty, then to hundreds. This role sits across the AI Toolbelt and Product Engineering lanes, owning the production AI systems that bank employees use every day — agent workflows, retrieval pipelines, and LLM integration layers. We bring a problem and expect a working solution.

WHAT YOU OWN

• Agent orchestration frameworks for multi-step reasoning, tool use, and constraint-based problem solving across banking workflows

• RAG pipelines covering embedding generation, chunking, hybrid retrieval, and retrieval evaluation, calibrated for banking document types

• LLM integration layers connecting banking models, APIs, and knowledge bases into reliable, auditable inference workflows

• Evaluation infrastructure including behavioral contracts, regression baselines, and production observability for non-deterministic AI outputs

• Backend services and APIs powering client-facing AI products at bank-tier uptime requirements

WHO YOU ARE

Background in software engineering with at least five years of experience, the last two spent building and operating production AI systems. Shipped agentic workflows, RAG pipelines, or LLM-powered applications to real users. Strong Python fundamentals across APIs and async systems, which is the foundation the AI work sits on. Comfortable picking the practical solution over the clever one.

Fluent in LangChain, LangGraph, PydanticAI, or AutoGen, with hands-on experience with vector databases, retrieval evaluation, and observability tooling such as LangSmith, RAGAS, Arize, or Langfuse. Prior fintech or banking experience is a genuine advantage, not a checkbox.

REQUIRED QUALIFICATIONS

• 5+ years software engineering; 2+ years building and shipping production agentic AI or RAG systems

• Agent framework experience: LangChain, LangGraph, PydanticAI, AutoGen, or Semantic Kernel

• RAG stack proficiency: embedding models, vector DBs (Pinecone, Weaviate, Milvus, FAISS), hybrid search, retrieval evaluation

• LLM integration depth: tool calling, structured outputs, multi-step reasoning, behavioral regression testing

• AI eval and observability tooling: LangSmith, RAGAS, DeepEval, Arize, Langfuse, or equivalent

• REST APIs, async Python, microservices; Azure cloud experience preferred

STRONGLY PREFERRED

• Fintech, banking, or regulated industry experience

• Graph databases (Neo4j, ArangoDB, Dgraph) and MCP / connector architecture

• Multi-agent or planner-based AI architectures

• Multi-tenant SaaS with auditability and compliance requirements

WHAT SUCCESS LOOKS LIKE

Within 90 days, ownership of at least one production AI workflow end to end with measurable improvements shipped to the retrieval or agent layer. Within six months, the go-to person on the team for hard agent and retrieval problems, operating independently from a high-level brief through to recommendation and implementation. At one year, a senior anchor on the AI engineering function with a track record of pulling others up and a credible path to leading other AI Engineers.

COMPENSATION AND STRUCTURE

• Competitive base and meaningful equity.

• Remote (US). Occasional travel to client sites and team offsites.