Welcome to the space of
Pranav Lokhande
Software Engineering Student and Research Assistant at University of Sydney
clicked by pranav
I'm a Software Engineering student at the University of Sydney, passionate about crafting accessible, performant web applications and creating thoughtful developer experiences.
I'm proficient in both frontend and backend development, from building clean, type-safe frontends in React and Next.js to designing robust APIs and data pipelines. Recently, I've been diving deep into LLM tooling and exploring pragmatic AI features that genuinely enhance user experience.
I'm currently contributing to automations in research projects and constantly explore systems design, product thinking, and visual polish.
I spend free time reading about design systems, experimenting with datasets, and refining workflows to ship faster with fewer regressions. When I'm not coding, you'll find me at the beach or playing cricket with friends.
Bachelor of Engineering (Software Engineering)
University of Sydney • Jun 2023 – Present
- • Engineering Data Science Specialisation
- • Peer mentor and student representative
Exchange Programs
International Universities • 2024 – 2025
- • NTU Singapore: Data Science, AI & Cybersecurity (Jun–Jul 2024)
- • Yonsei University: Big Data Analysis (Dec 2024 – Jan 2025)
- • SMU Singapore: Quantum Computing in Financial Services (Jun–Jul 2025)
Research Assistant
University of Sydney • Mar 2025 – Present
- • Researched and applied AI techniques to enhance educational resources and teaching tools.
- • Developed and fine-tuned LLMs for personalized learning and instructional applications.
- • Executed data preprocessing and model training to improve adaptive learning systems and efficiency.
Summer Researcher
Charles Perkins Centre, University of Sydney • Nov 2024 – Feb 2025
- • Developed & optimized data processing pipelines in Python using MzMine (Cu63, Cu65).
- • Created visualizations of isotopic peak distributions to support data interpretation.
- • Collaborated with Dr. Michael Gotsbacher and Dr. Lake-Ee Quek to integrate scripts into research workflows.
Predicting Building Energy Consumption – CNN, Bi-LSTM Hybrid Modelling
Machine Learning • Python, TensorFlow, PyTorch
Developing advanced deep learning models (Bi-LSTM and Transformer-based) to improve the accuracy of building energy consumption predictions, surpassing baseline metrics. Explored the effect of non-causal filtering to reduce noise and manage temporal dependencies, and conducted feature analysis to identify key predictors and optimize model performance.
TeamFlow – Full-Stack Project Management Tool
Web Application • React, Spring Boot, Supabase
Developing a full-stack web application to streamline task allocation and collaboration for student teams. Built with React on the frontend and Spring Boot (REST APIs, Spring Security) on the backend, using Supabase (PostgreSQL) for data storage and JWT authentication for role-based access control. Deployed via Vercel with CI/CD on GitHub.
Trains, Trends, and Turbulence – Data Analytics Project
Data Analysis • Python, pandas, Matplotlib, Folium
Analyzed the impact of major disruptions including COVID-19, large public events, and extreme weather on Sydney train ridership from 2020 to 2025. Leveraged Opal card tap-on data and BOM weather records, performing data cleaning, aggregation, and analysis in Python. Created visualizations and geospatial trend maps using Matplotlib, Seaborn, and Folium to identify commuter behavior shifts and quantify disruption impacts.
Fanum Tax – React & Supabase Tax Platform
Web Application • React, Supabase
Developed a secure platform simplifying tax management for individuals and businesses. Built with a React frontend and Supabase backend for authentication and database management, providing efficient, user-friendly tax filing and real-time support.
CollabEdit – C & Linux Real-Time Doc Editor
Systems Programming • C, Linux
Developing a collaborative document editing service in C on Linux. Implements a server-client architecture where multiple authorized clients can concurrently edit a Markdown-based document stored as plain ASCII text. Uses efficient linked list data structures to handle insertions and deletions without costly array shifting, ensuring the server always maintains the most up-to-date version.
Charles Perkins Centre Summer Research Scholarship
University of Sydney • 2024
Selected among 19 students for the Prestigious Summer Research Scholarship in collaboration with CPC.
Winter Data Analysis Challenge Honorary Mention
Sydney Precision Data Sci Centre & Westpac AU • 2024
Analyzed Sydney train ridership (OPAL) using Python (pandas, NumPy) and visualized trends with Matplotlib, Seaborn.