| Location | Cleveland, OH |
| Salary | $14.52–$15.40 Per Hour |
The Baker Nord Institute for the Humanities, Dr. Ben Vinson III (historian), and Mr. David Beales from the Freedman Digital Scholarship Center are hiring students with strong technology skills to develop and deploy a transformer-based OCR and text analysis pipeline, called BEN4.
The SOMOS project applies open source AI tools to make specific repositories within the Mexican National Archives machine readable. The project then analyzes the machine readable text with cutting-edge, best practices in text analysis.
Students will be hired as Class III student employees. Starting pay will be $14.52-$15.40/hr
The goal is to create a unique, adaptable, and highly effective digital/AI tool that can be used by humanists for any collection of interest. Accordingly, the project uses a modular design that allows different small models with context relevant training to be plugged into the pipeline as needs change.
The SOMOS project will offer opportunities to collaborate with partners in other institutions as we test models and work towards deploying the SOMOS-BEN4 pipeline in support of other collections.
We are looking for members to constitute the SOMOS team who have interest or expertise in some or all of the following areas:
Deploying open source AI/ML models using Ollama and Python
Optimizing systems for local hardware and cloud deployments
Evaluation and fine-tuning of LLMs
Computer vision, TrOCR, HTR and Image Preparation
GraphRAG, Topic Modelling, Named Entity Recognition and other text analysis methods
Web stack and UX/UI
Given the wide variety of experimental technological resources used in this project, an exceptional capacity for self-directed learning with technology is paramount.
Additionally, given that the historical collections we are working with are in Spanish, we are also looking for students who have intermediate to advanced reading knowledge of Spanish and demonstrate a willingness to learn colonial Spanish. It is not necessary, however, for students to have both the technological expertise and Spanish language skills.