Building
systems
at scale.

I designed and implemented an AWS Bedrock powered chatbot using Lambda and boto3 to deliver actionable insights and dynamically create AWS resources, including EKS, EC2, and S3. I also engineered an MCP server that connected Bedrock agents with Kubernetes APIs, enabling automated cluster management and operational support for over 150 teams. Additionally, I deployed Bedrock agents and action groups using Terraform, ensuring scalable, secure, and organization-wide accessibility across multiple AWS environments while maintaining consistent infrastructure-as-code practices.

At TD, I led efforts to streamline risk attribution processes by reducing over 14,000 risk mappings to just ~100 using regex-based pattern recognition and targeted data analysis. Building on this foundation, I developed a robust Spring Boot application to monitor more than 80 critical APIs across REST and ZeroMQ, delivering real-time visibility through a dynamic and actionable user interface. Furthermore, I also created a dashboard to compare trade metrics post curve/code changes to maintain accuracy of previous records. Finally, to ensure scalable and efficient communication across system components, I also utilized Protobuf, enhancing both performance and application modularity.

Participating in CodePath’s Technical Interview Prep (TIP) Course, a rigorous program focused on data structures and algorithms, including arrays, linked lists, stacks, trees, graphs, and dynamic programming. Solving Leetcode-style coding challenges, optimizing for time and space complexity. Engaging in mock interviews to refine problem-solving strategies and technical communication. Strengthening the ability to tackle complex problems, write efficient code, and approach technical interviews with confidence.

Engineered an Electronic Medical Records (EMR) system using Golang, available to rural hospitals in underprivileged areas. This will be pivotal in digitizing patient records (rather than being paper-based), making them increasingly accessible

Enhanced the US Navy’s PASS project with Machine Learning and Data Science to predict ship maintenance delays and optimize resource allocation. Improved model accuracy, reduced predictive errors, and delivered insights from classified data to senior officials, resulting in significant cost savings and better prediction accuracy.

Enhanced the analysis of PubMed data by applying NumPy to extract combinations from a vast number of citations, identifying various Social Determinants of Health (SDoH). Utilized NumPy and Pandas to refine the dataset and expand the SDoH ontology. Additionally, employed the UMLS API to gather synonyms from multiple databases, enriching the ontology with semantically similar matches for more comprehensive insights.

Developed discounted cash flow (DCF) models for a range of Fortune 500 companies to assess their financial health and investment potential. Conducted in-depth research across various industries to stay current with emerging trends and market dynamics, ensuring a comprehensive evaluation of investability and financial stability.