Kanupriya Guha
ML Engineer | GenAI & LLM Innovator | GPT-4, Claude, LangChain | MLOps & Cloud (AWS, GCP) | Turning AI Ideas into Scalable Products
Get in Touch
About Me
I'm Kanupriya a Machine Learning Engineer with over 4 years of experience turning complex data problems into real-world AI solutions. I've worked on everything from chatbots and document summarizers to large-scale LLM systems, and I love building things that make life easier for users and teams alike.
I'm always excited to explore how AI can solve meaningful problems—whether that's making customer support smarter, speeding up internal tools, or optimizing business processes.
Contact Information
📱 302-428-9926
📍 New Jersey
Professional Experience
1
Machine Learning Engineer
ServiceNow | September 2024 - Present
New Jersey, United States
2
Data Scientist
Neon IT Systems | October 2020 - July 2023
India
ServiceNow Achievements
Currently building enterprise-grade AI solutions that transform how teams work.
1
Generative AI Solutions
Designed and deployed enterprise-grade solutions using OpenAI GPT-3.5/4 and Claude for ticket summarization, document Q&A, and internal knowledge search.
2
LLM Pipelines
Built LLM-based pipelines with LangChain and FastAPI, enabling dynamic tool use, RAG workflows, and context-aware chatbot solutions with 90%+ response accuracy.
3
Model Fine-Tuning
Fine-tuned BERT and T5 models on domain-specific datasets, improving precision and reducing hallucination in generative tasks.
4
MLOps Excellence
Developed MLOps workflows with MLflow, Docker, and Kubernetes, enabling model monitoring, versioning, and rollbacks.
Neon IT Systems Impact
Key Accomplishments
  • Developed ML/DL models using TensorFlow, PyTorch, and Scikit-learn, including CNNs, Transformers, and RNNs
  • Led design of document summarization and Q&A engine using GPT-3 and Claude, increasing user search efficiency by 40%
  • Implemented sentiment analysis and NER using Hugging Face Transformers with over 90% accuracy
  • Built interactive dashboards with Tableau and Power BI for C-level strategy support
Technical Performance Metrics
90%+
Response Accuracy
Achieved in context-aware chatbot solutions using LangChain and RAG workflows
40%
Efficiency Gain
Increased user search efficiency through GPT-3 and Claude integration with Snowflake
20%
Forecast Improvement
Enhanced forecast accuracy for inventory optimization using RNN time-series models
4+
Years Experience
Building and deploying production-grade machine learning solutions
Core Expertise
LLM & GenAI
Built and deployed GPT-4, Claude, BERT, and LangChain-based applications for Q&A systems, chatbots, and document summarization.
End-to-End ML Pipelines
Experienced in taking projects from data wrangling to production using TensorFlow, PyTorch, Scikit-learn, and MLOps tools like MLflow, FastAPI, and Docker.
Cloud-Native Deployments
Hands-on with AWS SageMaker, EC2, GCP AI Platform, and container orchestration using Kubernetes.
Data Science & Analytics
Proficient in Python, SQL, R, and visualization tools like Tableau and Power BI for actionable insights.
Technical Skills
Programming Languages
Python, SQL, R for data analysis, model development, and statistical computing
ML/DL Frameworks
TensorFlow, PyTorch, Scikit-learn, Hugging Face Transformers, BERT, GPT-4, Claude
Cloud & DevOps
AWS SageMaker, EC2, GCP AI Platform, Docker, Kubernetes, MLflow, FastAPI
Data Visualization
Tableau, Power BI for creating interactive dashboards and executive reporting
LLM Tools
LangChain, RAG workflows, prompt engineering, fine-tuning, and model optimization
Data Engineering
Snowflake, data pipelines, ETL processes, and database administration
Key AI/ML Projects
Here are some of the impactful projects I've led and contributed to:
Designed and deployed an enterprise-grade chatbot using OpenAI GPT-4/Claude and LangChain, achieving 90%+ response accuracy for internal Q&A and support ticket summarization.
Developed a document summarization and Q&A engine using GPT-3 and Claude, boosting user search efficiency by 40% and reducing hallucination through fine-tuned BERT and T5 models.
Implemented a robust sentiment analysis and Named Entity Recognition (NER) pipeline using Hugging Face Transformers and Python, delivering over 90% accuracy for critical business insights.
Time-Series Forecasting Model
Built and deployed an RNN-based time-series forecasting model using TensorFlow and PyTorch, enhancing inventory optimization and achieving a 20% improvement in forecast accuracy.
Education & Certifications
Academic Background
Master's Degree
Rowan University
Data Modeling/Warehousing and Database Administration
August 2023 - June 2025
Bachelor's Degree
Mount Carmel College
Bachelor of Computer Applications (BCA), Computer Science
June 2019 - November 2022
Professional Certifications
  • Microsoft Excel Training
  • Graph Data Modeling Fundamentals
  • Importing CSV data into Neo4j
  • Introduction to Cassandra
  • Google Digital Garage: Fundamentals of Digital Marketing
Languages
English (Full Professional), Bengali (Native), Hindi (Professional), French (Elementary)
Let's Build Something Amazing
I'm passionate about leveraging AI to solve meaningful problems and create scalable solutions. Whether you're looking to implement cutting-edge LLM applications, optimize ML pipelines, or build cloud-native AI systems, I'd love to connect and explore how we can work together.
Ready to collaborate? Reach out today and let's turn AI ideas into impactful products.
Made with