Led an engineering team at Kunai in implementing observability tools for Capital One bank, utilizing New Relic and OpenTelemetry, and integrating AI for proactive incident detection. Instrumented multiple AWS-hosted microservices to improve real-time monitoring, traceability, and performance across diverse platforms. Integrated LangChain to develop RAG-based solutions for identifying incidents and recommending resolutions. Leveraged LLMs to detect duplicate incidents and accelerate issue resolution.

Designed and deployed an NFT Marketplace smart contract on the Avalanche network using Solidity, OpenZeppelin libraries, and Hardhat, allowing users to trade NFTs using ERC20 tokens and AVAX. Led development of ThorFi protocol V2 on Avalanche blockchain with innovative NFT-based model using ERC721. Developed multiple subgraphs to index transactions, trends, and analytics for ERC721 contracts on Avalanche. Automated CI/CD processes using GitHub workflows. Created microservice to index NFT metadata stored in AWS S3 and served via CloudFront. Led backend development using microservices architecture with Node.js and Docker.

Enhanced Conversational AI with ML Models using transfer learning on NLP models like GPT-2 for healthcare. Developed and deployed AI chatbot MVP for patient engagement and intent recognition. Created LENA Health Assistant platform integrating Patient Relationship Management and Case Management tools for personalized patient care.

Led development of Andia SDK's mobile framework for iOS and Android enabling digital KYC onboarding with biometrics. Developed machine learning models based on ResNet and DenseNet architectures for biometric identification (Face Recognition). Andia was part of Techstars WU 2019 cohort in Denver.

Designed and implemented text processing models for extracting relevant information from large text volumes. Created scalable backend for massive OCR and text processing tasks. Developed and managed RESTful APIs using Python and Tornado for ML model access.

Created and trained predictive forecasting model for foot traffic in retail stores. Developed high-performance platform processing millions of records per minute. Pioneered heat map solutions for foot traffic using WIFI-derived data.

Developed first earthquake alert app for Mexico City integrating public alert system to connect millions of smartphones. Designed high-availability platform supporting millions of users. Secured over 2 million active users and provided proactive alerts for over six major earthquakes with 50+ seconds advance notice.