Artificial Intelligence Engineer - Document Intelligence

About Company

Gloroots AI is a Y Combinator S24 startup dedicated to rebuilding freight finance from the ground up. We automate back-office accounting for freight brokers using advanced, freight-native AI agents that process complex accounting paperwork in seconds. Our founding team comprises leaders with extensive experience from Google, Georgia Tech, and Progressive, bringing deep AI and enterprise expertise to our mission.

Position Overview

As an AI Engineer specializing in Document Intelligence, you will play a pivotal role in designing and building intelligent agents that automate critical document-based freight workflows. Your work will directly impact our ability to process millions of freight documents, architecting core perception systems to extract structured insights from messy, real-world documents. You will also create AI agents for seamless email communications. Working closely with our experienced founders, you will have significant ownership over the AI stack and contribute to scaling our operations exponentially without proportional headcount increases.

Role & Responsibilities

• Develop accurate and reliable systems for extracting and analyzing knowledge from scanned, handwritten, and distorted freight documents.
• Architect, implement, and deploy AI agents for email and phone communications between freight accounting parties, leveraging LLMs and VLMs.
• Design and refine high-impact prompts, templates, and evaluation harnesses to ensure robust and reliable agent behavior.
• Build scalable pipelines for preprocessing, training, inference, and feedback loops, including VLM integration.
• Monitor and diagnose agent performance in production, rapidly addressing failures and refining models.
• Create and maintain high-quality training, evaluation, and test datasets for document intelligence workflows.
• Collaborate with product and engineering teams to integrate AI outputs into production-ready automation solutions.

Must Have Criteria

• 2+ years of hands-on ML/AI experience, with a preference for document AI or computer vision.
• Deep learning expertise with PyTorch or TensorFlow for model development and training.
• Proven experience crafting precise prompts for VLMs and LLMs, translating complex tasks into actionable instructions.
• Hands-on experience with OCR, visual transformers, and multimodal model architectures.
• Experience fine-tuning vision-language models for document understanding tasks.
• Proven track record of training and deploying ML models to production environments.
• Strong problem-solving mindset with the ability to prototype and iterate quickly in ambiguous environments.

Nice to Have

• Experience with AWS, Azure, or GCP-based ML infrastructure and deployment.
• Knowledge of document layout understanding models (e.g., Donut, LayoutLM, PubLayNet).
• Background in conversational AI, voice AI, or NLP research.
• Experience with RAG pipelines, foundation models, or vector search systems.
• Prior startup experience in fast-paced, high-ownership environments.

What We Offer

• Competitive salary with significant equity upside as an early-stage Y Combinator company.
• High ownership and zero bureaucracy, allowing you to shape the AI stack from day one.
• Opportunity to work on impactful real-world problems blending AI and automation at massive scale.
• Remote-first culture with flexible hybrid options.
• Direct collaboration with experienced founders from Google, Georgia Tech, and Progressive.