An enthusiastic Data Scientist with experience in Machine Learning, Natural Language Processing, Computer Vision, and Cloud Services. 6 years of work experience in Data Science, Site Reliability Engineering, and Web Development.
I believe that data can create wonders !!!
● Worked as a Machine Learning Engineer for an AI driven document search engine and GPT driven chatbots that provide information and answer questions about different departments in the organization.
● Built chatbots by integrating vector databases with ChatGPT using Databricks and Promptflow.
● Experimented with LLM models like ChatGPT, Mistral, LLaMA to reduce the cost by 20%.
● Involved in Prompt Engineering tasks for improving LLM’s answer correctness to 90%.
● Developed Chat PDF feature that generates ChatGPT responses from a given set of PDF files.
● Designed and implemented vector databases using Azure Cognitive Search and Weaviate for huge datasets.
● Performed Data Preprocessing, transformation using Pandas, Pyspark, Langchain.
● Automated RAG workflows using Databricks Asset Bundles which reduced 80% of the manual work.
● Worked as an AI/ML developer for a Document Intelligence based application that automates the process of creating the response documents for RFP (Request For Proposal) documents using AI/ML techniques which reduced the manual effort and time of SME’s by 20%.
● Built models for Custom SPACY NER to extract custom entities, Document Clustering, QnA for generating answers in a response document for the corresponding RFP document.
● Built NLP based models for Sentence similarity, text generation and text summarization using transformers, GPT.
● Created Databricks Jobs for performing AI operations using Azure Databricks. Refactored code to reduce the response time for the user by 40%.
● Designed and built the CI/CD pipeline using Azure DevOps. Written bash scripts for CI/CD automation.
● Created tables, views, user defined functions in Snowflake Cloud Data Warehouse.
● Extracted and loaded data from AWS S3 to Snowflake Cloud Data Warehouse.