OmniAI
Automate document workflows
Full Stack Engineer
About the role
Engineering at Omni!
We’re looking for founding engineers to join our team. Omni is right at the intersection of data engineering and LLMs. We’re looking for a great engineer who really want to join a startup.
What will we be working on
The main things we spend our time on:
- Wrangling LLMs into providing predictable outputs
- Running data transforms at scale
- Processing multimodal data (audio, documents, images, etc.)
All of these problems are hard, especially in conjunction with each other. If you’ve had any experience with structured LLM output, we’d love to chat.
Tech Stack
The primary tech stack is Node, TypeScript, React/NextJS, Postgres, Docker. For our integrations we support MySQL, Snowflake, Mongo, BigQuery and more. We don’t use these much internally, but our customers do so it’s helpful to know.
On the LLM side, we interface with OpenAI, Mistral, Llama, and Anthropic, so users have the choice of model to run.
Companies can use Omni via the cloud product, or a VPC deploy. So knowledge of Docker + devops is a big plus.
About the interview
Phone Screen
Take 30 minutes to chat with one of the founders. Learn a bit about your interests, and share what we’re working on at Omni. We are looking for people with a strong interest in startups, and would love to hear about prior startup experience. Also really interested in seeing any open source work.
Architecture Interview
A 1 hour call with both founders where we design a hypothetical feature. There’s no coding involved. Instead we’ll discuss a new feature, and do a collaborative walk though of the architecture required. We want to think about everything from database design to UI.
Onsite
You’ll spend the day working with us at our SF office. Meals will be covered, as well as any travel and accommodation. We’ll spend the day building out an MVP of the feature we planned out during the architecture interview. You'll be using your own computer, and using whatever language/environment you are most comfortable with. By the end of the day we are expecting two things:
- Fully functioning code. A few bugs are expected, but the core use case should be solved.
- A great demo. We want to see how clearly you can pitch your finished solution.
Mention logit biasing in your message if you read the JD.
About OmniAI
At Omni, we're building an entirely new way to work with unstructured data.
The most valuable data is often the least accessible. Only 20% of corporate data fits in a SQL table today, with the rest scattered across unstructured formats (customer reviews, chat logs, transcripts, PDFs, etc.)
With Omni, you can query and analyze that data as if it were SQL. We're enabling large scale unstructured data analytics that would have been impossible a few years ago.
