Karan Vijayvargiya
Senior Machine Learning Engineer @Tiger Analytics
Signup · Get unlimited contacts
WORK HISTORY
Senior Machine Learning Engineer @Tiger Analytics
Toronto, ON, CA
Developed an Agentic Chatbot powered by Langgraph React Agent that can answer questions asked by Insurance Policyholders to CSRs (Customer Service Representatives)- Deployed MCP (Model Context Protocol) server to retrieve policyholder data from AWS Redshift database in real time and integrate it with the Agentic Chatbot as a MCP Client. Leveraged FastMCP library to build the MCP server- Designed authentication flows (OAuth/JWT) to make sure the application stays secured and only the authenticated user is able to access- Worked in Agile environments using tools like JIRA, Confluence, and Git branching strategies for sprint tracking and collaboration. Applied rapid prototyping tools (Postman, Swagger, Streamlit) to visualize and validate ideas early- Developed technical documentation, architecture diagrams, and API usage guides to support internal teams and clients. Conducted hands-on workshops and knowledge-sharing sessions on ML model deployment, GenAI workflows, and MLOps best practices
EDUCATION
ACADGILD
Master's degree, Data Science
Birla Institute of Technology and Science, Pilani
Msc Physics(Hons.) with B.E. Mechanical(Hons.), Physics and Mechanical
Kendriya Vidyalaya
Inter, Science with Comp. Sc.
SKILLS
ABOUT KARAN VIJAYVARGIYA
I have always been fascinated by how data can tell powerful stories and drive impactful decisions. With over 10 years of experience spanning automotive, consumer appliance, and fintech sectors, I’ve had the opportunity to dive deep into the world of data science and machine learning. From developing real-time fraud detection systems at Moneris to automating anomaly detection using PCA & Deep Learning techniques such as Autoencoder at Infosys, each project has fueled my passion for creating solutions that make a difference.At Moneris, I pioneered a system that uses ensemble models to identify suspicious transactions seamlessly across card-present and e-commerce channels. This initiative not only decreased fraudulent activities but also enhanced customer trust—something I\'m incredibly proud of. Similarly, during my time at Whirlpool Corporation, I developed a machine learning model that achieved 97% accuracy in classifying fabric types—streamlining processes while reducing fabric damage.The thrill of working with cutting-edge technologies like Generative AI keeps me on my toes! I’ve gained a solid understanding of fine-tuning LLM models and implementing RAG based approach to achieve expected performance from LLMs. Integrating NLU into chatbots for tax products led to an impressive 90% accuracy rate in intent recognition—making tech accessible while enhancing user experience.If you’re looking for someone who combines technical expertise with real-world application—and who is equally passionate about solving complex challenges—I’d love to connect! Feel free to reach out via email at k••••••••@gmail.com, if you want to discuss collaborations or share insights about innovations in machine learning.Skills- Python, Pyspark, Databricks- Machine Learning- Deep Learning- Natural Language Processing- Generative AI- Fraud Detection- MLOps
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.