Software Engineer, AI (Senior/Staff)

Monarch is a leading personal finance platform aiming to simplify financial management for users. The role focuses on designing and building AI features that enhance user experiences by applying machine learning and AI technologies to real financial problems.

Responsibilities:

• Apply AI to Real Financial Problems: Use GenAI and ML to help users make sense of their money, whether that's understanding spending patterns, surfacing actionable insights, or automating tedious financial tasks. You'll identify where AI can make a meaningful difference and build solutions that users rely on daily
• Choose the Right Tool for Each Problem: Navigate the AI toolkit thoughtfully. Know when a well-crafted prompt suffices, when retrieval systems add value, and when custom models are worth the investment. You'll balance innovation with pragmatism to ship features that work reliably at scale
• Ship with Confidence: Leverage and enhance our sophisticated evaluation framework to ensure AI quality. You'll design test datasets, implement new scorers, and use our Braintrust-based eval system to validate changes before they reach users
• You Own: AI feature development, agent design and orchestration, ML model improvements, evaluation datasets and scorers, prompt engineering, and feature-level quality
• AI Platform Owns: LLM routing and provider management, observability and cost attribution, infrastructure reliability, and shared AI services
• Together You Own: End-to-end feature quality, evaluation frameworks, production incident response, and AI roadmap priorities

Requirements:

• 5+ years of experience in software engineering, with at least 2 years focused on building and operating production ML/AI systems
• A proven track record of shipping LLM-powered features, with deep, hands-on expertise in prompt engineering, RAG systems, and evaluation techniques
• Strong fundamentals in machine learning: embeddings, similarity search, classification, and probabilistic reasoning
• Demonstrated experience building and using AI evaluation tooling (e.g., golden sets, rubric scoring, LLM-as-judge)
• Excellent Python skills and a history of building production-grade AI features and services
• Strong collaboration and communication skills with a sharp product sensibility
• A strategic mindset, comfortable making build-vs-buy decisions and designing features for long-term reliability
• Multi-Agent Systems: Designing and building complex LLM orchestration with frameworks like LangGraph, CrewAI, or AutoGen
• Fine-Tuning: Hands-on experience with LoRA, RLHF, or full fine-tuning on platforms like Vertex AI
• Fintech Domain: Background in personal finance, banking, or data-rich consumer financial applications
• Vector Databases: Hands-on experience with OpenSearch, pgvector, Pinecone, or similar at scale
• Safety & Evaluation: Experience with red-teaming exercises, adversarial testing, and implementing guardrails