I build full-stack apps and train machine learning models—with a keyboard in one hand and filter coffee in the other.
I’m Milan Srinivas, a Software Developer pursuing my Master’s of Science in Computer Science at Northeastern University’s Khoury College of Computer Sciences. With hands-on experience across full-stack development, data science, mobile app development, and machine learning, I thrive at the intersection of code and creativity. I’ve worked across multiple startups, building scalable solutions and intuitive user experiences. I'm also passionate about deep learning and neural networks, with research work published in peer-reviewed journals. Explore my projects and feel free to connect with me on my socials!
Full Stack Developer – QSpiders Global, Bangalore, India Aug 2024 – Present
• Developed and maintained enterprise-grade full-stack web applications using Python, Django, and React, enabling seamless real-time frontend–backend interaction and reducing user-reported issues by 35%.
• Built responsive, cross-browser compatible UI components using TypeScript, Tailwind CSS, and HTML5/CSS3, enhancing design consistency and improving load times by 22% across client websites.
• Designed and optimized Oracle SQL and PostgreSQL databases, implementing normalization, indexing, and stored procedures to increase query performance by 40% for data-heavy applications.
• Automated CI/CD workflows via GitHub Actions, with secure versioning and rollback protocols; accelerated deployment cycles by 50%, adhering to DevOps best practices.
• Contributed to cloud infrastructure management using AWS EC2 and S3, achieving 99.9% uptime and improving recovery time objectives through effective monitoring and rollback strategies.
• Collaborated with clients for feature requirements and revisions, while assisting senior developers with large-scale code reviews and debugging across VS Code, Linux, and Git environments.
• Delivered mentoring sessions for over 50 industry professionals transitioning into tech, conducting workshops on web development, project management, and tools like Figma and GitHub to support their domain switch.
Software Development Intern – Konigtronics Pvt. Ltd., Bangalore, India Sep 2023 – Nov 2023 | Apr 2024 – Jun 2024
• Spearheaded data-driven sales analysis for a commercial automobile manufacturer, engineering scalable pipelines using Pandas, NumPy, and PySpark on Linux, reducing preprocessing time by 60% and improving data refresh frequency.
• Built and deployed Random Forest and Linear Regression models via Flask API and PostgreSQL, achieving 86.07% forecast accuracy, enabling data-backed decisions that led to a 23% increase in quarterly sales revenue.
• Delivered clear, actionable insights to non-technical stakeholders through interactive dashboards using Matplotlib, Seaborn, and Plotly; conducted presentations using MS Office tools in multiple boardroom meetings.
• Led a 10-member team in building a cross-platform mobile app using React Native, TypeScript, Tailwind CSS, and Firebase, integrating RESTful APIs and ensuring 100% device compatibility and zero post-deployment bugs.
• Followed Agile methodology with version control through Git/GitHub; interacted directly with the client to deliver iterative feedback-driven changes, improving app usability by 40% and achieving full client satisfaction within an 8-week timeline.
Software Development Intern – Contriver, Bangalore, India Jan 2024 – Mar 2024
• Collaborated directly with the CEO of a growing startup to build an end-to-end full-stack website for a virtual South Indian restaurant using HTML5, CSS3, JavaScript, and Python.
• Developed a responsive, SEO-optimized UI with CSS Grid/Flexbox and Vanilla JavaScript, boosting mobile performance by 35% and improving user engagement metrics.
• Integrated Google Maps SDK and external REST APIs to enable real-time dynamic content and geolocation features.
• Used WordPress and Python-based automation tools to streamline content updates and modern CMS practices.
• Deployed the site using GitHub Pages, CI workflows, and version control with Git/VS Code, ensuring cross-browser compatibility and maintainable codebase architecture.
Master of Science in Computer Science – Northeastern University 2025–2027
Bachelor of Engineering in Computer Science and Engineering – Visvesvaraya Technological University 2021–2025
CGPA: 8.7 / 10
Courses: Data Structures, Operating Systems, Object-Oriented Programming, Computer Networks, Machine Learning, Artificial Intelligence
Visual Question and Answering (VQA): ViT/SwinT and BERT/RoBERTA – IJARCCE
Explored the integration of Transformer-based models in Computer Vision (ViT/SwinT) and Natural Language Processing (BERT/RoBERTA). Developed a VQA system to analyze image-based questions, achieving improved interpretability and accuracy through multi-modal fusion.
A Hybrid Brain-Computer Interface and Computer Vision-Based Human-Vehicle Collaborative Driving System - IJIRCCE
Combined Brain-Computer Interface (BCI) and computer vision techniques to enhance collaborative autonomous driving. Integrated motor imagery and steady-state visual evoked potentials (SSVEP) with real-time road perception using OpenCV.
A Hybrid Reference Architecture Enabling Quantum Computing Capabilities for Cloud Utilization: Toward a Quantum-Science Gateway - IARJSET
Developed a hybrid architecture enabling quantum computing experiments on cloud platforms, bridging traditional and quantum computing models. Focused on streamlining access to quantum hardware for non-specialists, facilitating AI and financial applications.
Breast Cancer Histopathology Classification Using Levenberg–Marquardt Optimised Deep Neural Networks - IJARCCE
Engineered LMHistNet, a CNN model using Levenberg-Marquardt optimization and attention modules to classify breast cancer histology images. Achieved a 99% binary classification accuracy and 88% multiclass classification accuracy, addressing challenges like vanishing gradient.
Improving LLM-based Robot Control via Human-Robot Cooperation - IJIRCCE
Designed an LLM-driven robotic control system integrating natural language processing with real-time environmental perception. Enhanced motion planning algorithms to incorporate human inputs dynamically, improving task success rates to 91.1%.
