Senior Full Stack & AI Developer

Etsy

1. Developed a scalable e-commerce platform using React, Next.js, Nest.js, Python, PostgreSQL, and microservices architecture, while implementing an MLOps pipeline with Kubeflow, MLflow, and Seldon Core on Amazon EKS, resulting in a 35% increase in user engagement and streamlined machine learning workflows.
2. Created a Retrieval-Augmented Generation (RAG) chatbot leveraging LangChain and deployed Mistral 7B on Amazon SageMaker JumpStart using Hugging Face LLM DLC, powered by Text Generation Inference (TGI) on AWS, enhancing customer support interactions.
3. Built a recommendation engine using the Hypothetical Document Embeddings (HyDE) approach with Redis and the OpenAI API, increasing user engagement and sales.
4. Fine-tuned the Contrastive Language-Image Pre-Training (CLIP) model to improve item category and brand prediction accuracy, leading to more precise product classifications and an improved shopping experience.
5. Refined a Large Language Model (LLM) for a customer support chatbot using Quantized Low-Rank Adaptation (QLoRA) techniques, implementing canary deployments to ensure seamless updates.
6. Designed various prompt engineering strategies, including chain-of-thought and step-back prompting, to improve model performance and accuracy.
7. Customized Salesforce CRM to streamline customer data workflows, integrating marketing automation tools and enabling personalized campaigns, resulting in a 25% increase in customer retention.
8. Enhanced system security by implementing Role-Based Access Control (RBAC), integrating the Istio service mesh, and applying network policies within the Kubernetes cluster, ensuring compliance with best practices.
9. Devised and optimized continuous integration and deployment pipelines using Jenkins and Docker, reducing deployment times by 60%.
10. Migrated backend infrastructure to AWS, leveraging EC2 and S3 to reduce hosting costs by 30% while enhancing scalability.
11. Established automated testing frameworks using Jest, Mocha, and Cypress, along with A/B testing methodologies, achieving 99% test coverage and reducing production defects by 40%.
12. Integrated monitoring tools such as Datadog and Elasticsearch, enabling proactive bottleneck resolution and reducing system downtime by 30%, ensuring 99.9% platform uptime.

AI & Full Stack Developer

Token Metrics

1. Led the development of responsive user interfaces for a cryptocurrency analytics platform using TypeScript, React, Storybook, and Redux, contributing to backend API design with Ruby on Rails and MongoDB, and improving user engagement by 40% through innovative data visualization.
2. Created and maintained a vector-search-based recommendation system on AWS using Elasticsearch, resulting in an 18% increase in customer engagement and a 12% rise in sales.
3. Implemented a CI/CD pipeline for training and prediction using Kubeflow on Kubernetes to detect chargeback transactions in the Merpay payment service, employing natural language processing for transaction cancellation detection based on chat messages.
4. Enhanced MLOps pipelines for the deferred payment service with manual confirmation and implemented model management for client-side machine learning powered by TensorRT and Firebase.
5. Conducted neural architecture search (NAS) to automate multi-modal modeling for prohibited item detection.
6. Redefined and tracked KPIs for marketing initiatives using Tableau and SQL, implementing A/B tests that generated a 15% increase in client leads.
7. Reduced fraudulent activities by 49% by creating an anomaly detection system for financial transactions, safeguarding the company's assets.
8. Crafted custom data visualization dashboards using D3.js, and TensorFlow, integrating AI-driven real-time analytics to provide actionable insights and enhance decision-making capabilities for platform users.
9. Designed and deployed secure APIs and backend services, implementing OAuth authentication and integrating third-party blockchain data providers, improving data accuracy and reliability by 20%.
10. Automated the ETL process utilizing Apache Airflow, Kafka, Spark, and Tableau, improving data processing efficiency and scalability.
11. Built an AI-driven chatbot utilizing large language models to provide real-time cryptocurrency investment advice, leveraging financial models and market data to enhance user decision-making.
12. Streamlined deployment pipelines using Docker and GitHub Actions, reducing deployment times by 40% and minimizing production errors.
13. Integrated AWS services (S3 for storage, Lambda for serverless computing, and RDS for database management) to build scalable applications, reducing operational costs by 15%.

Full Stack Developer

Google

1. Engineered scalable microservices using Node.js and Java with Spring Boot to build a Global User Authentication System, leveraging Redis for caching and load balancers to ensure secure, low-latency access for millions of users with 99.99% uptime, improving system performance by 40%.
2. Architected highly interactive and accessible web interfaces for Google Workspace using Vue.js and Vuetify, creating dynamic dashboards and custom UI components that increased user engagement by 30% and improved enterprise workflow efficiency.
3. Designed and integrated RESTful APIs and GraphQL endpoints using Nest.js, implementing middleware for authentication and enabling cross-platform compatibility, reducing data retrieval times by 35%.
4. Implemented a hybrid backend system using Node.js for real-time processing and Java for transactional workflows, leveraging Kafka for message streaming and reducing processing latency by 50%.
5. Developed an AI-driven anomaly detection system using TensorFlow and Scikit-learn to enhance security in the Global User Authentication System, identifying fraudulent login patterns and reducing unauthorized access attempts by 35%.
6. Deployed scalable cloud applications using Kubernetes and Google Cloud Platform, utilizing Helm charts and auto-scaling to integrate Node.js microservices with PostgreSQL and BigQuery, enhancing scalability and reducing operational costs by 20%.

Software Engineer

Accenture

1. Contributed to multiple client projects, including developing a real-time logistics tracking system, a secure payment service, and a healthcare analytics dashboard.
2. Engineered dynamic, pixel-perfect, user-friendly web interfaces using React and Angular, collaborating with designers in Figma to translate wireframes into high-quality web pages.
3. Constructed scalable backend systems with Node.js, Ruby on Rails, C#, and Python, incorporating PostgreSQL and MongoDB for efficient data storage.
4. Enhanced performance and user experience by implementing SSR, SSG, and PWA techniques, improving SEO visibility, boosting search rankings, and reducing server load by 35%.
5. Optimized frontend performance with advanced caching techniques, prefetching, and lazy loading using React tools, reducing page load times by 50% and increasing user retention.
6. Streamlined development processes using Webpack, Babel, and NPM for optimized code bundling and cross-browser compatibility, while collaborating with teams via Git, GitHub, and Bitbucket.

Columbia University

Computer Science