Backend/AI Engineer

About the role

You will own the design and delivery of the backend systems that power our agentic document platform at enterprise scale. You will be responsible for integrating cutting-edge large language models and building the infrastructure that turns unstructured documents into reliable, production-grade data. This role demands a high bar for quality, where you act as your own worst critic and refuse to ship anything less than a fully finished and robust solution. You will work directly with founders and customers to translate ambiguous problems into scalable engineering outcomes, moving fast with high agency. A core part of your ownership will be building internal tools and evaluation frameworks that give us deep insight into model behavior and failure cases. Ultimately, you will make state-of-the-art document understanding accessible to hundreds of companies through a performant and reliable API.

Key facts

What you'll do

Integrate and optimize LLM calls for structured extraction, form filling, and advanced document understanding tasks across diverse document types.

Make iterative improvements to API design and document preprocessing algorithms such as chunking and structured extraction based on direct customer feedback and usage data.

Build and enhance document processing pipelines that reliably handle the complexity of PDFs, spreadsheets, and other unstructured formats at massive scale.

Experiment with novel output structures, prompt strategies, and techniques to boost LLM accuracy while simultaneously reducing system latency and cost.

Construct internal tooling and evaluation suites to analyze failure cases, monitor model performance, and drive continuous improvement in extraction quality.

Work closely with founders and enterprise customers to shape product direction, translating high-level requirements into concrete technical specifications and priorities.

Develop quantitative approaches to building and testing products, using data to guide hypotheses, debug issues, and iterate rapidly from idea to production.

Build and maintain robust integrations with leading frontier models, ensuring that the platform leverages the best available capabilities for document understanding.

Create streamlined developer experiences and workflows that allow our AI teams to prototype, test, and deploy new document intelligence features with high efficiency.

Champion software craftsmanship by writing clean, maintainable code and rigorously reviewing contributions to ensure the stability and performance of core services.

Requirements

You have 2+ years of experience building real world applications and integrating LLMs into production systems, with a proven track record of shipping reliable software.

You are exceptional at Python, writing idiomatic and efficient code that is easy to maintain and scale in production environments.

You possess a quantitative approach to building products, using experimentation and data to drive decisions and validate hypotheses quickly.

You have a strong bias for action, preferring to ship fast with high agency and taking ownership of problems end to end without waiting for direction.

You are comfortable getting hands-on with the full development lifecycle, from initial ideation and prototyping through to deployment, monitoring, and iteration based on user feedback.

You have a high bar for quality and do not rest until the job is done right, refusing to accept technical debt or partial solutions.

You actively collaborate with cross-functional teams, communicating clearly with founders, product managers, and customers to align on goals and constraints.

You are comfortable working in a fast-paced, early-stage environment where responsibilities evolve rapidly and flexibility is essential.

Nice to have

Prior experience founding a company or building products at early stages, understanding the constraints and opportunities of a startup environment.

Hands-on experience with prompt engineering, fine-tuning, or building AI agents to solve complex tasks using language models.

A strong commitment to keeping up with the latest developments in machine learning and artificial intelligence to bring fresh ideas into the product.

Practical notes

This is an in person role at our office in SF. We're an early stage company which means that the role requires working hard and moving quickly. Please only apply if that excites you.