Senior ML Engineer

Saris AI•San Francisco, CA

About The Position

Saris AI is an applied AI startup building the future of work in the banking industry. We focus on automation problems requiring long-context reasoning, tool orchestration, and agentic decision-making in a highly regulated sector. We have successfully deployed AI agents handling real customer workflows in production and are seeking to expand our engineering team to scale our existing solutions and innovate further. The core engineering team is looking for a Senior ML Engineer who excels in early-stage, ambiguous environments. This role involves building and owning the ML infrastructure for reliable and improvable AI systems, including evaluation frameworks, prompt management, and model observability. The engineer will also ship customer-facing AI features, define the team's approach to evaluations, LLM routing, prompt engineering, and model selection, and establish quality standards without hindering progress. Additionally, the role requires contributing to ML technical direction by identifying trade-offs and architectural options to guide informed decision-making.

Requirements

4+ years of experience in ML or AI engineering, with a track record of shipping production ML systems
Strong hands-on expertise with LLMs, prompt engineering, evals, and model routing
Experience building tooling and systems that have real customer impact
Pragmatic about tradeoffs: knows when good enough is the right call and avoids over-engineering; would rather ship something useful today than design something perfect next quarter
Comfortable working with moderate direction in ambiguous environments, you can take a scoped problem, work through it, and deliver a shipped solution
Builds with the end user in mind; understands how ML decisions impact real customers and prioritizes customer value over technical elegance
Elevates teammates through code review, pairing, and clear communication about technical decisions

Nice To Haves

Worked in regulated industries (fintech, banking, healthcare) where compliance and reliability are first-class concerns
Experience with RAG systems, fine-tuning, or open-source LLM deployment alongside closed models
Comfortable across the stack, data pipelines through APIs, and can plug gaps where needed
Used or built prompt management or ML observability tooling

Responsibilities

Build and own the ML infrastructure that makes our AI systems reliable and improvable, including eval frameworks, prompt management, and model observability
Ship customer-facing AI features on a consistent cadence, balancing new capability delivery with foundational infrastructure work
Define and implement the team's approach to evals, LLM routing, prompt engineering, and model selection
Build pragmatic standards that improve quality without slowing the team down
Contribute to ML technical direction by proactively surfacing trade-offs and architectural options, helping the team make informed decisions on where ML is headed