Senior AI Engineer – Vision

This is a fully remote position, open to applicants in Latin America.

📋 Description

• Unlocking Visual Data: Developing pipelines capable of interpreting intricate documents, including layout, charts, and visual context, using Vision-Language Models (GPT-4V, Claude 3.5) and Layout Analysis.

• Orchestrating Intelligence: Taking ownership of the application logic layer. You will utilize LangChain or LangGraph to construct agents and chains that query our data, reason with it, and generate responses.

• Native PDF Handling: Navigating the complexities of PDF processing (PyMuPDF, layout parsing) to maintain structural integrity before AI analysis.

• Prompt Engineering & Logic: Designing intricate prompts and control flows to guarantee accurate interpretation of financial charts and layouts by the models, minimizing hallucinations.

• Cost & Scale: Implementing a cost-optimization approach (batch processing, model selection) to ensure our vision and orchestration layers remain economically feasible.

⛳️ Requirements

• LLM Orchestration (Must Have): Extensive experience with LangChain, LangGraph, or comparable frameworks. You possess knowledge in managing context windows, tool invocation, and agent-based workflows.

• Multimodal AI Experience: Practical experience in integrating cutting-edge vision models (GPT-4V, Claude 3.5 Sonnet) and embedding models (CLIP).

• Document Intelligence Specialist: Acquainted with specialized models (e.g., Donut, Pix2Struct) and tools such as Unstructured.io or Docling.

• PDF Processing Mastery: Expert-level familiarity with tools like PyMuPDF or pdfplumber for native element extraction.

• Python ML Stack: Strong skills in PyTorch or TensorFlow.

🏝️ Benefits

• 18 days of PTO per year, observance of local holidays, and an annual break between Christmas and New Years.

• A monthly wellness stipend and snack boxes delivered to your home.