Full Stack Engineer

Omni is building the best OCR / document extraction platform on the planet, focusing on transforming unstructured data into queryable SQL-like formats. As a founding engineer, you'll be tackling challenging problems at the intersection of document processing and AI.

Key Focus Areas:

1. Wrangling LLMs to provide predictable outputs
2. Scaling document extraction systems
3. Building training data for vision models
4. Working with their open-source library (github.com/getomni-ai/zerox)

The role involves working with a modern tech stack including Node.js, TypeScript, React/NextJS, and Postgres. You'll also work with various integration points including MySQL, Snowflake, MongoDB, and BigQuery. On the AI side, you'll interface with multiple LLM providers including OpenAI, Mistral, Llama, and Anthropic.

This is a unique opportunity to join a early-stage startup tackling the challenge of making unstructured data (which makes up 80% of corporate data) accessible and queryable. You'll be working directly with the founders to build and scale the core product, which can be deployed either as a cloud solution or within customer VPCs.

The role requires strong full-stack development skills, experience with modern web technologies, and ideally some familiarity with Docker and DevOps practices. Experience with structured LLM outputs would be particularly valuable, though not required.

As an early team member, you'll have significant input into technical decisions and architecture, while working on problems that combine document processing, machine learning, and scalable web services. The position offers both technical challenges and the opportunity to shape the direction of a promising startup in the document processing and AI space.