Software Engineer & CP Enthusiast
JAY
DHANWALKAR
Transforming complex algorithmic challenges into elegant, scalable full-stack solutions.
Technical Arsenal
Technologies and tools I use to build things.
Experience & Leadership
- Managed the full Software Development Life Cycle (SDLC) to design and deploy custom web applications using Next.js and TypeScript.
- Optimized performance, achieving 95+ Lighthouse score and 30% faster load times.
- Automated CI/CD with GitHub Actions & Vercel.
- Integrated secure auth & PostgreSQL ensuring data integrity.
Technical Member
Student Club, GHRSTUFeb 2025- Led a CP workshop for 300+ students (Recursion, Graphs, Trees).
- Spearheaded "Frontend Face-off" competition for 100+ participants.
Featured Works
AI Chat Application
Built a scalable AI-driven chat interface utilizing the Gemini AI API to provide context-aware, real-time responses. Analyzed user query patterns to refine prompt engineering, increasing response accuracy by 50%.
Discuss Design Develop
Developed a real-time video conferencing platform featuring screen sharing and collaborative whiteboarding. Engineered a shared interactive canvas with real-time state synchronization.
Sorting Visualizer Tool
Implemented tool for visualization of various sorting algorithms like quick sort, merge sort etc. at various size and speed. Prioritized a user-friendly design resulting in a 30% increase in task completion efficiency. Built on HTML, CSS And JavaScript.
Sales Kart
Experienced full-stack developer skilled in creating seamless eCommerce websites using cutting-edge technologies. Proficient in backend development with Node.js and Express.js, paired with MongoDB or PostgreSQL for efficient data management. Adept at building responsive frontend interfaces with React.js, employing state management techniques like Redux for optimal performance.
Sudoku Solver
Developed a highly efficient Sudoku solver emphasizing advanced data structures and algorithms. Achieved a 20% improvement in solving speed and 15% fewer backtracks. Conducted extensive testing across various Sudoku puzzle scenarios, achieving a 98% success rate.
Hackathon Achievement
OpenEnv National Hackathon
Meta • PyTorch • Hugging Face • X • Scaler School of Technology
Built an RL environment where Qwen2.5-0.5B learns to fix Python bugs using test-execution feedback. Implemented procedurally generated tasks covering diverse bug types (index errors, syntax, division by zero, None handling). Trained with GRPO using multi signal rewards for robust learning.Achieved +400% reward improvement (0.2 → 1.0+) over 80 training steps with 5/6 tasks fully solved.
Milestones & Academics
Meta Hacker Cup
Achieved All India Rank 6262 in the global competition orchestrated by Meta.
723+
DSA Problems Solved
MCA
G.H. Raisoni College, GHRSTU
-
Google Cloud Agentic AI Day 2025: Participant and Hack2Skill attendee.
-
Microsoft Learn Student Ambassador: Selected virtually (2024).
-
GDSC WOW Event: Participant in Google Developer Student Clubs event (Mar 2022).
-
Google Kick Start: Participant in the global online coding competition (Mar 2021).
BCA
Sardar Patel College - Gondwana University