Software Engineer (LLM Products)

Blue Rose Research•New York, NY

10h•$110,000 - $210,000•Remote

About The Position

Blue Rose Research is seeking a product-focused engineer to help build a new generation of AI-powered products, including conversational interfaces, autonomous research agents, and tools that transform large datasets into actionable information. The role involves collaborating with internal and external users and the political research team to understand their needs, translating these insights into product ideas, UX designs, and production code. This position is ideal for a builder-designer comfortable with both design tools like Figma and coding. The ultimate goal is to contribute to Democratic electoral success in upcoming elections. The work environment is fast-paced, with evolving priorities and often ambiguous problems. The role is primarily remote, though the company has an office in NYC and facilitates in-person meetups in DC. Most work is conducted on East Coast time.

Requirements

6+ years of experience as a product engineer, full-stack engineer, or designer-engineer hybrid.
A portfolio of shipped products.
Strong product and UX instincts: ability to critique a flow, sketch an alternative, and explain tradeoffs.
Comfort using coding agents to build quickly without compromising quality.
Hands-on experience building products on top of LLMs: chat interfaces, agents, retrieval-augmented tools.
Understanding of how to use human annotations efficiently to encode human expertise into AI systems.
Energized by ambiguous problems.
Authorization to work lawfully in the United States.

Nice To Haves

Familiarity with tools like Figma.
Motivation by Blue Rose Research’s mission and interest in learning the progressive political ecosystem.
Prior political or civic-tech experience.

Responsibilities

Design and build end-to-end features for products where an agent and humans are working together.
Build products that present AI-generated output, uncertainty, and citations in a way people actually trust.
Work closely with political researchers, data scientists, and other engineers to turn ambiguous problems into simple, usable software.
Use AI tools and coding agents to move fast, and know when a slicker interface actually matters versus when it’s polish nobody needs.
Talk to users and internal stakeholders to understand what’s confusing, slow, or untrustworthy about existing tools, and fix it.
Dive head first into the world of political strategy to build systems that encode a deep understanding of the user’s goals and thought processes.