Priya Bhardwaj

Results-oriented Python Developer with expertise in building scalable applications, optimizing system performance, and leveraging machine learning models. Proficient in Python, Django, and data visualization, with hands-on experience in delivering user-centric solutions. Adept at data analysis, algorithm design, and problem-solving. Passionate about driving innovation through impactful, data-driven technologies. Recent graduate with excellent research, technical, and problem-solving skills. Detail-oriented and able to learn new concepts quickly.

• Developed scalable, modular code structures, reducing debugging and update time by 20%.
• Improved application response times by an average of 25% through performance optimization.
• Enhanced data analysis processes with NumPy and Matplotlib, enabling precise data-driven decisions.

Find Movie Similarity from Plot Summaries 

• Tech Stack: Python, Tkinter, NLP (TF-IDF, Cosine Similarity)
• Developed a content-based filtering model for movie recommendations, increasing search relevance by 25%.
• Designed a Tkinter-based interface, improving user navigation and engagement by 20%.
• Achieved response times under 2 seconds for recommendations, enhancing user satisfaction.
• GitHub Repository: Movie Similarity from Plot Summaries

Meal Deal Suggestion App  

• Tech Stack: Python, Machine Learning
• Built a machine learning model for personalized meal deal recommendations, achieving a 90% match accuracy rate.
• Fine-tuned algorithms to enhance recommendation accuracy, increasing app engagement by 15%.
• GitHub Repository:Meal Deal

Hide and Seeker: Multi-Agent Reinforcement Learning Simulation   

• Tech Stack: Python, Q-Learning, SARSA, TensorBoard
• Developed a 5x5 grid-world with two Hiders and a Seeker using Q-Learning and SARSA to optimize the Seeker's strategy for capturing the Hiders. Implemented reward shaping to encourage efficient learning, achieving faster convergence with Q-Learning and safer exploration with SARSA.
• Conducted experiments over 2,500 episodes, with exploration decay and optimized learning parameters.
• GitHub Repository: Hide and Seeker Project