Generative AI Engineer

NovaSpark AI

Remote · $105K – $125K · Full-time · Posted 6 months ago

This role is no longer accepting applications

Don't miss the next role like this

Create a free profile with your skills. We match you to new AI roles as they open, and companies hiring can contact you directly.

Create free profile, 2 minutes

Similar open roles you can apply to now:

See all open AI jobs

This role is no longer accepting applications

Don't miss the next role like this

Create a free profile with your skills. We match you to new AI roles as they open, and companies hiring can contact you directly.

Create free profile, 2 minutes

Similar open roles you can apply to now:

See all open AI jobs

About the role

We got into YC six months ago. We launched three months ago. We have 11,000 users and $0 in VC funding beyond our YC check, which means every decision we make has to be a good one. Here's what we're building: a legal document assistant that helps small business owners understand contracts without paying $400/hour for a lawyer. Our early users love it. Our current stack is OpenAI + LangChain + a retrieval layer that's frankly too slow and too expensive to scale. That's the problem we need you to help solve. We're not looking for someone to maintain what we have. We're looking for someone who will look at our architecture and tell us — honestly, specifically — what's wrong with it and what to do instead. We're two engineers and one designer. You'd be the third engineer. You'd have real ownership. You'd work on problems that matter to people who genuinely can't afford alternatives. If that sounds like a good way to spend the next year or two, let's talk.

Fully remote.

Responsibilities

Audit and rebuild our document retrieval pipeline to reduce latency and per-query cost by at least 40%
Implement chunking, embedding, and re-ranking strategies appropriate to legal document structure
Own prompt architecture end-to-end: system design, A/B testing, and evaluation
Help define our technical roadmap and make the case for or against architectural decisions in writing
Work directly with users to understand failure modes and translate them into engineering improvements

Requirements

3+ years of software engineering with at least 1 year building LLM-powered features in production
Hands-on experience with LangChain, LlamaIndex, or equivalent RAG frameworks — not just tutorials
Strong Python: async patterns, clean architecture, you can write code other people maintain
OpenAI and Anthropic API experience including tool use, function calling, and streaming
Vector databases in production: Pinecone, Weaviate, Qdrant, or similar
A genuine point of view on what makes RAG systems fail at scale

Benefits

Real equity in a company with paying users and genuine traction — not a paper promise
Full remote, async-first culture, no mandatory stand-ups
$105,000 – $125,000 base salary
$1,000 equipment stipend on day one
You will have a meaningful say in what we build and how we build it