Junior AI Developer / Student

CBCL Limited

1d•Onsite

About The Position

CBCL is seeking a Junior AI Developer – Document Intelligence & Local Model Deployment to help design, build, and maintain a self-hosted AI platform that extracts structured information from complex technical documents and automates quality assurance workflows. The platform runs entirely on local GPU hardware using open-source models. This is a greenfield build where you will be standing up local inference infrastructure, building document processing pipelines, and delivering a working prototype on real project data within your first term. You will work directly with experienced engineers and technical staff who will provide domain expertise, validation feedback, and real-world test cases. This position is a 4-month term with the possibility of extension based on project milestones and mutual fit. CBCL is actively investing in applied artificial intelligence to improve how we deliver engineering services. Our AI development is done in-house, on our own infrastructure, using open-source tools — an approach driven by our commitment to client data sovereignty and our belief that the firms who build their own capabilities will lead the next generation of engineering practice.

Requirements

Enrolled in or recently completed a degree in Computer Science, Software Engineering, or a related field (senior undergraduate, M.Sc., or equivalent practical experience)
Strong proficiency in Python
Hands-on experience deploying and running open-source language models or vision-language models locally (e.g., local inference servers, quantized model deployment, or self-hosted model serving)
Experience processing PDFs or other document formats programmatically
Comfortable working in Linux (Ubuntu) from the command line
Working knowledge of Git for version control and collaborative development

Nice To Haves

Experience building context retrieval or semantic search pipelines (vector search, embedding models, retrieval strategies)
Familiarity with computer vision or object detection workflows
Experience with web application development (Python backend and/or modern JS frontend)
Exposure to GPU configuration and model optimization
Familiarity with some combination of the following open-source tools and frameworks: Ollama, vLLM, LangChain, LlamaIndex, Hugging Face Transformers, PyMuPDF, Tesseract, OpenCV, Ultralytics, Detectron2, FastAPI, Streamlit, Gradio, ChromaDB, FAISS, Label Studio, Docker
Any experience with engineering drawings, construction documents, CAD, or technical documentation of any kind

Responsibilities

Assist in the design and development of internal AI tools and models supporting engineering projects
Deploy and configure open-source AI models on local GPU infrastructure running Linux
Build a document ingestion and processing pipeline for structured and unstructured technical content
Build context retrieval pipelines that give the model access to domain-specific reference material at inference time
Design and iterate on prompt templates that produce consistent, structured outputs from the model
Work with engineers and technical staff to validate model outputs against real project data — you’ll learn the domain from the people who know it
Maintain clean, well-documented code in Git from day one
Document your architecture decisions, model configurations, and pipeline logic so the system is reproducible and maintainable