Large Language Models (LLMs) Engineer in GlobalLogic

Description:

Document Completeness Checker: Ensuring that all uploaded customer documents are complete and requesting any missing reference documents necessary for parameter identification.
Expanded Document Constructor: Developing comprehensive versions of both customer requirements and reference project documents by integrating information from referenced documents, including nested references.
Deviation Summarizer: Creating initial summaries of key differences between customer requirements and reference project documents, highlighting critical deviations.
Electrical Component Identification: Utilizing AI-driven object identification algorithms to identify and list electrical components in customer SLD diagrams.
Parameter Extraction & Mapping: Extracting and associating relevant parameters such as dimensions and material properties from customer specifications with the identified electrical components.
User Interface: Developing a user-friendly interface for engineers to review and edit the list of electrical components and corresponding parameters, with future integration with 3D tools in the MVP phase.
Identical Term Presence Parser: Checking for elements missed in the Costing Sheet, focusing on identical terms used across the DoW and Costing Sheets.

Requirements:

3+ years of experience in machine learning and natural language processing.
Proficiency in training and fine-tuning Large Language Models (LLMs).
Experience with RAG architectures and vector databases like Pinecone.
Bachelor’s degree in Computer Science, Engineering, or equivalent.
Excellent problem-solving and analytical skills.
Demonstrated ability to conduct experiments and improve model performance.
Good command of English for effective communication with team members.

Knowledge of Pytorch, Pinecone, LLM fine-tuning, RAG, and vector databases.
 

Job Responsibilities:

Train and fine-tune Large Language Models (LLMs) for various applications.
Implement and optimize Retrieval-Augmented Generation (RAG) architectures.
Work with vector databases like Pinecone to enhance data retrieval capabilities.
Develop and apply data science and machine learning techniques.
Collaborate with cross-functional teams to integrate LLMs into products.
Stay updated with the latest advancements in LLM and machine learning technologies.
Conduct experiments and analyze results to improve model performance.