Shivam is a Data Engineer with 5+ years of experience in building and optimizing data pipelines, deploying AI/ML projects, and integrating various cloud services. He has a strong background in Python, SQL, and AWS, and has successfully led teams in developing innovative solutions.
Developed and deployed an AI Voice Assistant on AWS EC2, capable of scheduling and answering product queries.
Led the development of a PDF summarization ChatBot using Langchain and OpenAI, streamlining information extraction.
Built efficient and robust ETL pipelines for large datasets, significantly enhancing data ingestion and processing efficiency.
Successfully deployed an AI Voice Assistant on AWS EC2.
Automated various data pipelines using Python, reducing manual intervention and improving data flow efficiency.
Implemented a Car Recommendation system utilizing collaborative filtering and Random Forest to predict hot-selling vehicles.
Overview: Built an AI voice assistant to answer queries and manage schedules for users, serving as a platform for AI-driven customer interaction. Responsibilities: Gathered data from PDFs to form the knowledge base for the organization. Responsible for integrating RAG (Retrieval-Augmented Generation). Built the general architecture diagram for the system. Integrated Twilio with AI for communication functionalities. Retrieved data from GHL (GoHighLevel) to create contacts, appointments, tasks, and notes. Fine-tuned the LLM model and evaluated its performance using a BLEU score. Deployed the entire project on an AWS EC2 instance.
Key outcomes:
Successfully deployed an AI voice assistant on AWS EC2.
Fine-tuned an LLM model for specific business query resolution.
Overview: Developed a ChatBot leveraging Langchain and OpenAI models to summarize PDF data, streamlining information extraction for efficiency. Responsibilities: Used the OpenAI API key to import Langchain for AI model integration. Utilized Pinecone as a VectorDB for efficient data retrieval. Employed OpenAI Embeddings for finding similar searches by vectorDB. Responsible for integrating RAG (Retrieval-Augmented Generation) within the chatbot. Led the project with a team of 6 developers, orchestrating the integration of the Langchain framework with OpenAI models for PDF data summarization.
Key outcomes:
Successfully developed an innovative ChatBot for PDF summarization.
Led a team of 6 developers, ensuring seamless integration of AI frameworks.
Overview: Developed an ML model for sales prediction to provide insights for inventory management, revenue growth, and new investment decisions. Responsibilities: Gathered requirements directly from the Business Team and collected necessary data. Performed Exploratory Data Analysis (EDA) to understand data patterns. Applied various Feature Engineering and Feature Selection techniques.
Key outcomes:
Built and validated an ML model that provided actionable sales predictions.
Overview: Developed a system to recommend hot-selling cars to users visiting a website, predicting cars likely to sell within 24-48 hours. Responsibilities: Created the ETL pipeline using AWS Glue, connecting to MongoDB and Apache Kafka. Connected to Databricks to perform machine learning operations. Applied feature transformation and filtering, then stored the processed data into a Data Warehouse.
Key outcomes:
Successfully developed a predictive car recommendation system.
Shivam
Data Engineer