ApprovalMax is award-winning B2B software used by businesses around the world to approve bills and expenses. It streamlines the approval process for decision-makers by replacing paper and email approvals with automated workflows. ApprovalMax integrates with platforms such as Xero, QuickBooks Online, and Oracle NetSuite to unlock powerful efficiencies for approvers and finance teams. The company is globally recognized and has won ’Xero App Partner of the Year’ four times since 2020 in different countries.

We're looking for a Senior Machine Learning Engineer to join our AI team at ApprovalMax. You'll design and build agentic AI systems that automate complex approval workflows for finance teams across 17,000+ businesses. This is a high-impact role where your work directly shapes how companies handle millions of financial transactions.
What You'll Do

• Design, build, and deploy agentic systems using LangGraph and LangChain for workflow automation

• Develop and maintain ML models for anomaly detection, fraud prevention, and intelligent document processing

• Build evaluation frameworks and implement observability using LangFuse

• Deploy and scale ML services using Ray Serve or FastAPI + Redis + Celery

• Work with vector databases (pgvector) for RAG-based retrieval systems

• Collaborate closely with product and engineering to ship AI features with high accuracy requirements

• Contribute to architectural decisions for our growing AI infrastructure

What You Bring
Required:

• 4+ years of production ML experience

• Strong Python skills with strict typing practices

• Hands-on experience building LLM-powered applications (agents, RAG, prompt engineering)

• Experience with at least one agent framework (LangGraph, LangChain, or similar)

• Solid foundation in classical ML (classification, regression, anomaly detection)

• Experience deploying ML models to production (Ray Serve, FastAPI, or equivalent)

• Familiarity with async processing patterns (Celery, Redis, or similar)

Nice to Have:

• Experience with Azure cloud services (Azure OpenAI, Container Apps)

• Background in fintech or financial document processing

• Experience with ML observability tools (LangFuse, LangSmith, Weights & Biases)

• Knowledge of vector databases and embedding models

• Experience with evaluation frameworks for LLM applications

Tech Stack

• Languages: Python

• Agent Frameworks: LangGraph, LangChain

• Deployment: Ray Serve, FastAPI, Celery, Redis

• Observability: LangFuse

• Infrastructure: Azure (Container Apps, OpenAI Service)

• Data: PostgreSQL, pgvector

What we offer:

• Growing international business with 10,000+ subscribers

• Regular performance-based compensation reviews

• 26 days paid time off

• 1 additional day off for your Birthday

• Remote office assistance

• Service years recognition financial reward