LLM Orchestration at Able

Setup: Remote
Location: Miraflores, Department of Lima
Type: Full-time
Posted: May 29, 2026

Description

This position involves leveraging Computer Vision and Logic to extract complex data from visual documents and integrate it with Large Language Models. The ideal candidate will build pipelines for document analysis, orchestrate intelligence using LangChain/LangGraph, handle PDF processing, craft precise prompts, and optimize costs. This role requires deep experience with LLM orchestration, multimodal AI integration, document intelligence specialists, and Python ML stack proficiency.

Required Skills

Computer Vision
Logic
Vision-Language Models
Layout Analysis
LangChain
LangGraph
PDF Processing
Prompt Engineering
Cost Optimization
Multimodal AI Integration
Document Intelligence
Unstructured.io
Docling
PyMuPDF
pdfplumber
PyTorch
TensorFlow

Additional Information

Required Previous Experiences: Deep experience with LangChain, LangGraph, or similar frameworks.
Hands-on experience integrating state-of-the-art vision models (GPT-4V, Claude 3.5 Sonnet) and embedding models (CLIP).
Familiarity with specialized models (e.g., Donut, Pix2Struct) and tools like Unstructured.io or Docling.
Mastery over tools like PyMuPDF or pdfplumber for native element extraction.
Strong proficiency in PyTorch or TensorFlow.
Preferred Qualifications: Experience fine-tuning vision or language models, specifically to improve accuracy on domain-specific artifacts like financial charts or tables.
Prior experience handling documents in the Real Estate or Finance sectors.