Meet the Founder
Paper Mentor AI was founded to make high-quality AI research mentorship more personal, practical, and technically rigorous.
Odysseas Drosis
Founder & Lead Mentor · Machine Learning Engineer at Meta · Cornell University
Odysseas is a machine-learning engineer and research mentor
with experience guiding students through applied AI projects
in healthcare, finance, sports analytics, computer vision,
natural language processing, and other machine-learning domains.
He currently works as a Machine Learning Engineer at Meta
and has an academic background in computer science and artificial
intelligence from Cornell University.
Students he has personally mentored have developed complete
research projects and published work in research journals
and student research publications.
Meta, Cornell University, Stanford University, and other institutions referenced on this website do not sponsor, endorse, or operate Paper Mentor AI. Institutional names are used only to describe professional, educational, or publication backgrounds.
Mission & Philosophy
Our goal is to bridge the gap between classroom learning and genuine research. Instead of simply following tutorials or completing a standard coding project, students learn how to define a research question, work with real data, design experiments, evaluate results, and communicate their findings clearly.
- Understand how modern machine-learning methods work in practice.
- Develop a focused research question around a real problem.
- Implement and evaluate models using real-world datasets.
- Compare approaches, interpret results, and identify limitations.
- Develop a structured research paper or technical report.
- Explore appropriate publication, competition, or research opportunities.
The emphasis is on understanding, experimentation, and independent thinking — not simply producing a finished paper.
What Makes the Mentorship Different
Paper Mentor AI is built around focused 1:1 work rather than large classes or pre-recorded courses.
1:1 Mentorship
Each research plan is adapted to the student's interests, experience, pace, and goals.
Research First
Students learn how to formulate hypotheses, evaluate models, analyze evidence, and think critically about results.
Hands-On AI
Projects involve real implementation, datasets, experimentation, and technical problem solving.
How Mentoring Works
Mentoring is typically organized as a sequence of individual online sessions, with project work completed between meetings.
- Identify an AI area that matches the student's interests.
- Review relevant research and narrow the project into a clear question.
- Build the required programming and machine-learning foundations.
- Implement models and run experiments.
- Analyze results and refine the methodology.
- Structure, write, and revise a research paper or technical report.
- For suitable projects, explore relevant journals, competitions, or other research outlets.
The technical depth and project scope are adjusted to the student's current level. Advanced programming experience is not required for every project.
Student Research & Publications
Students personally mentored by the founder have worked on research involving cancer detection, brain imaging, stock-price forecasting, sports analytics, NLP, and other applied AI topics. Selected projects have progressed to publication in research and student research journals.
View Published Student Research Explore ProgramsInterested in Starting a Research Project?
Tell us about the student's interests, experience, and goals. We will review the application and suggest what type of AI research project may be a strong fit.
Apply for Mentorship