Tech Stack

Description

You will work on building intelligent financial assistant systems that help users manage their finances and interact with financial products. You'll design agentic architectures, evaluation infrastructure, and safety guardrails while owning production observability. This is a customer-facing role in a regulated environment where accuracy and trust are critical.

Requirements

• Strong classical ML fundamentals
• Strong eval mindset: design measurement before prompts
• Hands-on experience with search and retrieval (dense/sparse/hybrid, reranking, query understanding)
• Practical experience with agentic architectures
• Expertise in Python and ecosystem, language- and framework-agnostic mindset

Responsibilities

• Work with agentic architectures: tool use, multi-step reasoning, orchestration, failure recovery
• Design and maintain eval infrastructure (offline test suites, golden datasets, regression harnesses)
• Work with advanced prompts, set up RAG-oriented data sources, evaluate model outputs
• Implement safety and guardrails: hallucination detection, refusal strategies, factual grounding
• Run online and offline experiments (canary, A/B) and own observability for LLM calls, quality, latency, cost