AI Engineer

About Humata

Humata is a seed-stage startup based in Austin, Texas. We are first to market with our Chat with Documents app and are building a state-of-the-art GenAI system to power document- and knowledge-base-question-answering. We are looking for an AI Engineer to help build the cognitive architecture that combines GenAI, Search, and Retrieval-Augmented Generation (RAG).

As an AI Engineer at Humata, you will work at the forefront of GenAI and information retrieval, building innovative architectures that push the boundaries of what’s possible with document and knowledge-base question-answering. You will have the opportunity to:

• Invent and Implement: Prototype and ship new methods of document search including image-based RAG, document understanding, and ML-based re-ranking.
• Own and Impact: Take end-to-end ownership of important projects as Humata builds out the cognitive architecture.

• 5+ years of software engineering experience
• Proven production experience with LLMs (e.g., GPT-4 or Llama 3.1)
• Expertise in Retrieval-Augmented Generation (RAG) with demonstrated success in tuning applications for accuracy
• Strong knowledge of search engines and information retrieval (e.g., Elastic Search, Vespa)
• Experience with cloud platforms (AWS, GCP) and SQL databases
• Proficiency in Typescript and Python

• AI/ML research background with experience in training and fine-tuning LLMs
• Experience with Node.js development

• Ownership Mentality:
  Takes initiative and drives projects to completion with minimal oversight.
• Scrappy and Action-Oriented:
  Thrives in a fast-paced, early-stage environment, delivering frequent, impactful code changes.
• Collaborative Team Player:
  Values team health, open communication, and collective success.

$150,000 - $200,000 a year

Innovate: Push the boundaries of GenAI using both closed- and open-source tools.

Ownership and Impact: Shape critical systems and processes in an early-stage startup.

If you’re an engineer passionate about building world-class AI systems and eager to tackle the challenges of Retrieval-Augmented Generation, we’d love to hear from you! Apply today to be part of our journey to redefine how users interact with documents and knowledge bases.