Local AI-Powered Contract Review System Developed for the Maritime Industry
A local AI pipeline has been developed that analyzes demurrage clauses in maritime contracts without cloud infrastructure while preserving privacy.
Stephen Day III and Wathila Ekanayake have developed a secure AI RAG pipeline that analyzes complex clauses and demurrage conditions in maritime transport contracts entirely locally, without the need for cloud systems.
Solution Within the Scope of the Thesis
Prepared as a thesis project by Stephen Day III and Wathila Ekanayake under the guidance of Dr. Elena Dugundji, the study aims to accelerate contract evaluation processes in maritime transport.
Privacy Needs and Local Infrastructure
Due to the high confidentiality requirements of maritime contracts, the system operates entirely locally without using any cloud infrastructure or third-party data centers.
Hardware and Memory Capacity
Running on a Minisforum MS-S1 Max workstation and an AMD Ryzen AI Max+ 395 processor, the system allocates 96 gigabytes of memory at the hardware level for AI workloads.
Model and Interface Features
The system simultaneously runs a 26-billion parameter language model and an 8-billion parameter embedding model, offering a web dashboard that facilitates the work of analysts.