Jessica López Espejel
Position
Deep Learning Researcher at Novelis
Professional Summary
Jessica López Espejel is a seasoned Deep Learning Researcher specialized in Artificial Intelligence (AI) with a focus on Natural Language Processing (NLP). She has a strong academic background with a PhD in Artificial Intelligence from Sorbonne Paris Nord University and CEA-LIST, France. Her research interests include Large Language Models (LLMs), text and code generation, and automatic evaluation.
Current Role and Responsibilities at Novelis
- Title: Deep Learning Researcher
- Tenure: September 2021 – Present
- Responsibilities:
- Develop and apply techniques for automated code generation in Java and SQL.
- Fine-tune models, optimize parameters, and engineer prompts for superior performance.
- Implement deployment of models for efficient execution.
- Generate production-ready optimized code.
- Publish scientific papers to validate proposed solutions.
Notable Achievements and Publications
Jessica’s research contributions are well-recognized with multiple publications focusing on advanced NLP techniques and AI applications. Some highlights include:
1. GPT-3.5, GPT-4, or BARD? Evaluating LLMs reasoning ability in zero-shot setting and performance boosting through prompts (2023).
2. A comprehensive review of State-of-The-Art methods for Java code generation from Natural Language Text (2023).
3. JaCoText: A Pretrained Model for Java Code-Text Generation (2023).
4. GeSERA: General-domain Summary Evaluation by Relevance Analysis (2021).
Academic Background
- PhD in Artificial Intelligence (2021)
- Sorbonne Paris Nord University and CEA-LIST, France
- M.Sc., Computer Science and Mathematics (2017)
- Benemérita Universidad Autónoma de Puebla, Mexico
- Engineering in Computer Science (2015)
- Benemérita Universidad Autónoma de Puebla, Mexico
Previous Positions
Postdoctoral Researcher
- Institution: Université Sorbonne Paris Nord
- Tenure: May 2021 – August 2021
- Responsibilities:
- Identified the existence of diasporas using reinforcement learning.
- Wrote optimized, clear code using best practices.
Researcher Ph.D. Student
- Institutions: Université Sorbonne Paris Nord & CEA-LIST
- Tenure: March 2018 – May 2021
- Responsibilities:
- Proposed methods for automatic summary generation.
- Developed techniques for automatic summary evaluation.
- Provided Python classes to Bachelor’s students at ENSTA Paris.
Software Developer
- Company: Totalplay Telecommunications
- Tenure: July 2017 – February 2018
- Responsibilities:
- Specialized in Java web services (front-end/back-end).
- Implemented security algorithms for webpage protection.
Professional Interests
- Artificial Intelligence (AI)
- Large Language Models (LLMs)
- Code/Text Generation
- Automatic Evaluation
- Natural Language Processing (NLP)
- Brain-Computer Interface (BCI)
Professional Networks
- LinkedIn: [Jessica López Espejel - LinkedIn](https://fr.linkedin.com/in/jessicalopezespejel)
- Google Scholar: [Google Scholar Profile](https://scholar.google.com/citations?user=aJ5A8hsAAAAJ&hl=es)
- ResearchGate: [ResearchGate Profile](https://www.researchgate.net/profile/Jessica-Lopez-Espejel)
- X (formerly Twitter): [Twitter Profile](https://twitter.com/jessica_11101)
Notable Presentations
- Maximizing Model Usability in Industry through Pruning: An Essential Optimization Technique
- Event: 4th Edition of the Data Science School in Benin, Africa
- Date: November 7, 2023
Research Contributions and Impact
Jessica has made notable contributions in the field of AI with 97 citations and a significant focus on the comparative analysis of open-source language models and Java code generation from natural language text. Her work is substantiated through rigorous peer-reviewed publications and ongoing involvement in advancing AI technologies at Novelis.
