About me

My name is Poshak, and I am a Computer Science major with a strong interest in data analytics, statistical modeling, and sports performance analysis. I am currently focused on developing my skills in extracting insights from data using tools such as Python, R, SQL, and Tableau, with experience in libraries like Pandas, scikit-learn, ggplot2, and dplyr.

This website is a space to document my analytics projects which are related to machine learning applications, data visualization techniques, and sabermetrics.

I am actively seeking opportunities to contribute to data-driven teams and continue learning from real-world challenges. Feel free to explore my work, and don't hesitate to connect with me on LinkedIn

What i'm doing

  • data icon

    Data Analytics

    I analyze real-world datasets using Python, R, SQL, and Tableau, with a focus on sports and research data.

  • ml icon

    Machine Learning

    I apply ML techniques like neural networks, random forests, and GANs to real-world problems including federated learning and sports forecasting.

Resume

Education

  1. University of Louisiana Monroe

    Aug 2022 – May 2026

    B.S. in Computer Science and B.S. in Mathematics. Active in GDSC, ACM, and ULM Math Club. Focused on applied machine learning and data analytics.

Experience

  1. Undergraduate Research Assistant – ULM

    Jan 2024 – Present

    Developed a GAN-based detection method for poisoned data in federated learning. Built data pipelines and trained ML models, improving accuracy by 25%.

Research Publications

  1. Pathak, L. S., Pathak, P., & Mishra, R. K. (2025). Immersion in second language does not always attenuate and attrite first language: Evidence from Nepali–English bilinguals.

    International Journal of Bilingualism

    Investigated and cross-checked if immersion in a bilingual situation over a certain period of time had any modulatory effect on bilingual processing in an immigrant population.

  2. Pathak, P., Sreekumari, P. (2025). Adaptive GAN-Based Defense Against Data Poisoning in Federated Learning

    ITNG 2025. Advances in Intelligent Systems and Computing, vol 1463.

    Proposed a GAN-based framework for real-time detection of poisoned data in federated learning. Efficient against targeted and untargeted attacks.

  3. Pathak, L. S., Vulchanova, M., Pathak, P., & Mishra, R. K. (2024). Trilingual parallel processing: Do the dominant languages grab all the attention?

    Bilingualism: Language and Cognition

    Published peer-reviewed research on cognitive load and attention in trilingual speakers, with implications for educational design.

  4. Pathak, L.S., Rijal, S., & Pathak, P. (2021). Instruction in second language enhances linguistic and cognitive abilities in first language as well

    J Cult Cogn Sci, Vol 5

    Studied public school children in Nepal and showed positive linguistic transfer from second to first language acquisition.

Other Achievements

  1. ULM Chess Club – Founder & President

    2023 – Present

    Established and led the university's official chess club. Drafted the club's constitution, organized biweekly events, and facilitated member growth through tactics, strategy, and classic game analysis.

  2. Hackathons

    2023 – 2025

    Participated in iQuHack 2023 (MIT), TAMU Datathon 2024 (Texas A&M), and won 2nd Place at Hacklytics 2025 (Georgia Tech).

  3. Conference Presentations

    2024 – 2025

    Presented research at the Louisiana Academy of Science (2024), ULM Research Symposium (1st Place), and ITNG Conference – Las Vegas (2025).

Portfolio

Contact

Contact Form