I build intelligent systems for AI, data, and products.
Engineer. Builder. Thinker.
I'm Rishi, a Software Engineer at Wells Fargo with a passion for bridging the gap between state-of-the-art AI research and reliable production systems.
My expertise lies in building intelligent backend systems. I specialize in developing RAG-based AI chatbots, optimizing LLM workflows for operational efficiency, and designing data-intensive pipelines using Java (Spring Boot) and Python.
Graduating from NIT Karnataka, I combine strong algorithmic foundations with practical experience in fraud detection models, cash prediction algorithms, and full-stack development. I believe that the best systems are not just "smart" but are also predictable, observable, and genuinely useful.
Core Fluency
Model Architecture
Development
Infrastructure
Developing RAG-based AI chatbots to automate banker workflows, improving operational efficiency by 26%. Designed ATM cash prediction models with 92% accuracy and built real-time data ingestion pipelines using Spring Boot and Grafana.
Designed and implemented a Personalized Fraud Detection System. Scaled backend infrastructure, trained high-accuracy ML models on customer data, and deployed RESTful APIs for fraud detection microservices.
Developed a Python-based recommendation algorithm leveraging user skills and past experiences to suggest relevant gigs. Proactively identified and reported product bugs to enhance user experience.
Major GPA: 8.39/10.00. Minor in School of Management. Published research on 'Image Manipulation Detection Using Augmentation and Convolutional Neural Networks' (MDCWC 2023).
Building intelligent systems using RAG, Chatbots, and LLM automation involved in the agentic workflow.
From idea → MVP → scalable systems. Full-cycle development with a focus on robust architecture.
Leveraging financial data and tech skills to build practical tools that simplify day-to-day life.
Interactive learning platform that teaches Git through hands-on tutorials and practice playgrounds. Features step-by-step lessons, real examples, and a safe environment to experiment.
Browser extension that automatically backs up coding practice submissions from LeetCode and GeeksForGeeks to GitHub.
An exploration of why true randomness in music shuffling often feels clumpy to humans, and how Spotify engineered a 'random' shuffle that actually feels random.
A deep dive into how Uber engineered a real-time catalog search system capable of handling billions of records with low latency using Apache Pinot.
A personal reflection on the Financial Independence, Retire Early (FIRE) movement, and why the traditional math might overlook crucial economic factors.
Software is my craft, but it's not my whole world. Here’s what keeps me grounded, creative, and curious offline.
Capturing moments with a low-budget phone. It forces me to focus on composition and lighting rather than relying on expensive gear.
I love all genres—from classical to techno. I may be a 'lyric killer' who can't remember the words, but the rhythm and melody always stick.
Understanding human and animal behavior. It's fascinating to see how our minds work and why we do the things we do.
Observing how geopolitics and supply & demand shape markets. I enjoy analyzing how the stock market reacts to global events.
High-altitude trekking and climbing. Finding clarity where the air is thin helps me reset my perspective on complex problems.
Capturing moments with a low-budget phone. It forces me to focus on composition and lighting rather than relying on expensive gear.
I love all genres—from classical to techno. I may be a 'lyric killer' who can't remember the words, but the rhythm and melody always stick.
Understanding human and animal behavior. It's fascinating to see how our minds work and why we do the things we do.
Observing how geopolitics and supply & demand shape markets. I enjoy analyzing how the stock market reacts to global events.
High-altitude trekking and climbing. Finding clarity where the air is thin helps me reset my perspective on complex problems.