Fuse Finance - Founding AI Engineer

Company Overview

Fuse is a high-growth New York City startup revolutionizing financial services infrastructure through an innovative low-code tool. Backed by top-tier venture capital, the company is creating a transformative self-service platform for configuration, integration, and automation in financial technology. We are now expanding our capabilities to leverage artificial intelligence in revolutionizing how financial services workflows and integrations are built.

Role Overview

We’re hiring a Founding AI Engineer to spearhead our AI agent initiatives and accelerate AI product MVPs to real clients. You’ll combine strong fullstack engineering (TypeScript) with LLM Ops to design, build, and deploy prototypes that validate product–market fit quickly—while establishing rigorous evaluation and human-in-the-loop safeguards for quality and trust. You’ll collaborate with Product Operations, coordinate two junior/mid engineers, and work with Design to deliver end-to-end—from ideation to production.

Core Focus (What Matters Most)

  • Rapid AI Prototyping & MVP Delivery: Design, build, and ship MVPs to clients to validate hypotheses and speed time-to-market.

  • Pragmatic Full-Stack + Low-Code: Enhance our low-code platform by generating/optimizing workflows, UIs, and data schemas powered by AI agents.

  • Performance Metrics & Human-in-the-Loop: Define and monitor accuracy, confidence, and error rates; create evaluation frameworks and set confidence thresholds that trigger human review.

  • End-to-End Ownership: Act as the MVP’s “CEO/CTO”—align with client needs, deploy, learn from real usage, and iterate aggressively.

  • Cross-Functional Leadership: Partner with Product/Engineering and Design to integrate AI seamlessly and hit aggressive timelines.

Responsibilities

  • Build and optimize AI agent pipelines leveraging leading LLMs and AI services to extend our low-code platform

  • Develop agentic workflows that ensure high-quality answers with strong domain grounding.

  • Generate and validate use-case-specific workflows, UIs, and data schemas.

  • Create evaluation frameworks to measure and improve output quality.

  • Integrate AI capabilities into the existing platform architecture and establish monitoring, versioning, and continuous improvement practices.

Must-Have Qualifications

  • 8+ years in fullstack engineering with proven, hands-on AI agent workflow implementation.

  • Expertise in TypeScript/Node.js, LLM integration (GPT, Claude, Gemini, Grok), and prompt engineering.

  • Experience with agent evaluation, pipeline design, and prompt/version control.

  • Strong foundations in RESTful APIs, data modeling, architectural patterns, and authN/authZ for AI environments.

  • Track record of production-grade AI systems delivering measurable business value.

Technical Expertise

  • LLMs & LLMOps; agent evaluation/optimization with tools like Deepchecks, OpenAI Evals, LangSmith (or similar).

  • AWS, SQL/optimization, ETL, RESTful API integration/management/design, and software development best practices.

Ideal Project Experience

  • Fine-tuning LLMs; agent-centered workflow pipelines; generative AI applications; automated testing/validation; documentation and knowledge sharing.

Technical Environment

  • Languages: TypeScript / Node.js, Low-code tools (Loveable, n8n)

  • AI/ML Frameworks: e.g., LangChain, LlamaIndex, Haystack, Flowise

  • Cloud: AWS

  • Database: PostgreSQL

  • Version Control: Git

  • CI/CD: GitHub Actions.

Nice to Have

  • Python (PyTorch, TensorFlow, Pandas or similar).

Apply for this job