Atharva Khaire

- Develop a real-time question-answering system using WebSockets (FastAPI), LangGraph agents, and transformer-based models, enhancing user interaction through intelligent query handling. - Lead a team of 10 engineers, overseeing project timelines, sprint planning, and task assignments using JIRA, while fostering collaboration and delivering scalable AI-driven solutions. - Design and implement a Retrieval-Augmented Generation (RAG) pipeline with LangGraph, LangChain, and Pinecone vector databases, advancing LMS capabilities for seamless knowledge retrieval. - Build APIs with LangChain, integrated with OpenAI APIs and Hugging Face models, to optimize user LinkedIn profiles using advanced NLP and embedding techniques.

- Deployed Airflow on AWS EKS using Docker, Terraform, and Jenkins, incorporating PySpark for distributed processing and Prometheus for monitoring, achieving 20% higher deployment frequency and 99.9% uptime. - Developed a scalable data pipeline using MongoDBAtlasVectorSearch and Hadoop for large-scale storage, optimizing vector indexing and improving query efficiency of jobs data by 50% through advanced caching and partitioning. - Engineered a Kafka-based event streaming system for real-time data ingestion into MongoDB microservices, utilizing Protobuf for message serialization, optimized Kafka consumer groups and MSK triggers, SQL for data transformations, and partitioning strategies, enhancing processing speed by 30% and ensuring data accuracy. - Designed and managed ETL workflows, integrating data from SQL Server into data lakes and warehouses using Airbyte and Fivetran, while enforcing ACLs, resulting in a 30% increase in data integration efficiency.

• Leading Assignment design for Association Rule Mining, Kmeans and Dimension reduction topics. • In charge of formulating questions for Midterm and Final Exams. • Organizing Office hours every week, for student performance improvement and ensuring clarity of concepts. • Responsible for grading assignments and exams for 120+ students.

DAG Migration: Spearheaded a comprehensive DAG migration initiative to EKS and Databricks, enhancing system agility and scalability by 25% using DBT models. PySpark Optimization: Authored over 1,000 lines of PySpark code, boosted data processing efficiency by 25%. Kafka Pipeline Development: Developed Kafka pipelines for credit card data transfer using financial web services to the Snowflake Data Warehouse, resolving 50+ critical issues in PySpark and SQL, enhancing data integrity by 20%. Business Intelligence Systems: Designed and maintained integrated business intelligence systems using Looker and data warehousing principles, employing LookML views and models, improving data accessibility efficiency by 30%.

• Providing academic support to students enrolled in CS 677 A1 under Professor Eugene Pinsky. • Leading review sessions and providing feedback on assignments and exams. • Assisting instructor with course planning and preparation. • Holding office hours to answer student questions and provide additional assistance. • Demonstrating a strong understanding of course material. • Possessing excellent communication skills and the ability to work well with students and instructors.

• Optimized Spark-based ETL pipelines, reducing data processing time by 35% for real-time material tracking. • Integrated IoT sensors with AWS IoT Core, Lambda, improving operational efficiency by 20% with predictive insights.