👋 Hello, I'm

Atharva Deshmukh

Data Scientist & AI Context Engineer

3+ years building production GenAI systems and multi-agent workflows. Expert in Model Context Protocol (MCP), LLM fine-tuning, and statistical ML, delivering 70-90% efficiency gains across healthcare and enterprise domains.

Atharva Deshmukh
3+
Years Experience
10+
Projects Delivered
2
Cloud Certifications

About Me

I am a Data Scientist and AI Context Engineer with 3+ years of experience building production GenAI systems, multi-agent workflows, and machine learning solutions that deliver measurable business impact.

My expertise spans Model Context Protocol (MCP), LLM fine-tuning, and statistical ML across healthcare, enterprise, and environmental domains. I specialize in designing AI agents, RAG systems, and deep learning models that have achieved 70-90% efficiency gains in production environments.

Currently completing my Master's in Information Technology and Analytics at Rochester Institute of Technology, I combine academic rigor with hands-on experience in MLOps, cloud platforms (AWS, Azure, GCP), and cutting-edge GenAI technologies to solve complex real-world problems.

MS
IT & Analytics
BE
Computer Eng.

Data Science

Statistical analysis, predictive modeling, and data visualization expertise

Machine Learning

Deep learning, neural networks, and ensemble methods for complex problems

Generative AI

LLMs, RAG systems, and multi-agent AI architectures

Professional Journey

Data Scientist (AI/ML)

CrowdDoing

Sep 2025 - Present
  • Built a wildfire satellite image-classification model using PyTorch (Landsat-8 + MODIS), achieving 87% precision and 82% recall, reducing false alerts by 30% across monitored regions.
  • Developed a real-time inference pipeline on AWS SageMaker, combining environmental sensor signals with ensemble ML models to prevent early wildfire spread.
  • Integrated LLM-based natural language search (LangChain + OpenAI) into dashboards for conversational access to risk insights, improving coordination efficiency by 25%.
  • Built MCP-powered AI agent for automated QA with council-of-agents validation, dynamically generating Playwright tests and reducing manual testing time by 70% while improving coverage to 95%.
  • Built self-healing MCP agent system with dynamic context window management and automated error recovery, reducing context overflow issues by 90% and enabling seamless handling of 1000+ daily agent interactions.
PyTorch AWS SageMaker LangChain MCP Playwright

AI/ML Engineer

CVS Health

Jan 2025 - Sep 2025
  • Designed a clinical risk-prediction model for medication non-adherence, improving recall from 71% to 84%, supporting reduction in preventable readmissions.
  • Fine-tuned clinical embeddings with LoRA (Hugging Face) to improve medical case-classification accuracy by 19%, enabling more reliable treatment prioritization.
  • Built FastAPI inference microservices on Vertex AI for real-time decision support used by pharmacy and case-management teams.
  • Implemented MLflow for experiment tracking, version control & CI/CD integration, reducing troubleshooting and release cycle time by 30%.
  • Delivered Power BI dashboards and automated reporting pipelines to visualize patient risk factors and enable data-driven decisions for healthcare leadership.
LoRA Hugging Face FastAPI Vertex AI MLflow

ML Engineer

VR Digital Solutions

Mar 2021 - Jul 2023
  • Designed and deployed machine learning models for customer segmentation, churn prediction, lead scoring, and campaign optimization, improving marketing ROI and targeting efficiency by 17% across multiple client accounts.
  • Built computer vision pipelines for product-image tagging and defect detection using PyTorch and TensorFlow, reducing manual QA inspection effort by 40% for e-commerce clients.
  • Engineered ETL workflows using Python, SQL, Pandas, Airflow to clean and transform 3M-8M daily records, improving data processing time by 2.1x and reducing reporting delays.
  • Developed REST inference APIs using Flask + Docker to integrate ML models with internal business dashboards enabling automated analytics and real-time predictions for sales teams.
  • Implemented MLOps practices including experiment tracking, training pipelines, and model versioning using MLflow + Git to enable reliable deployment and reproducibility.
PyTorch TensorFlow Flask Docker Airflow

Technical Skills

Programming

Python
SQL
R
Java

ML/DL Frameworks

TensorFlow
PyTorch
Scikit-learn
Keras

Cloud Platforms

AWS
Azure
GCP
OCI

Data Tools

Pandas NumPy Matplotlib Seaborn Tableau Power BI

Gen AI & LLMs

LangChain LlamaIndex OpenAI API Hugging Face RAG Prompt Eng.

MLOps & Tools

Docker Git MLflow Airflow Kubernetes CI/CD

Featured Projects

Agentic AI

LLM Based Resume Scorer

AI-powered multi-agent system leveraging LLMs to optimize resume–job description matches for improved ATS screening and candidate evaluation.

LangChain OpenAI Python
View Project
Agentic AI

Vizard AI

Intelligent AI platform using LangChain to automatically generate business-relevant data visualizations and insights from raw datasets.

LangChain Streamlit GCP
View Project
Machine Learning

LendSafe

AI-powered loan decision explanation system with FCRA-compliant reasoning using fine-tuned IBM Granite 350M for transparent lending decisions.

IBM Granite LoRA Fine-tuning Gradio
View Project
Agentic AI

Hamcaller Custom LLM

Custom fine-tuned LLM model designed to identify and filter spam calls using advanced NLP techniques to protect users from fraud.

Ollama Fine-tuning NLP
View Project
Agentic AI

Stars Yapp

Generative AI-powered astrology predictor delivering personalized insights using LLMs and RAG architecture for context-aware predictions.

RAG LLMs FastAPI
View Project
Data Analysis

Formula 1 Telemetry

Advanced analytics on F1 telemetry data to model driver performance patterns and build predictive insights for race outcomes.

R ggplot2 Analytics
View Project
Machine Learning

ML on AWS Sagemaker

End-to-end ML pipeline to predict phone prices using ensemble methods, deployed as scalable REST API on AWS Sagemaker.

AWS Sagemaker XGBoost
View Project

Certifications

Microsoft Certified: Azure Data Scientist Associate

Credential ID: C2F98ADE9E9ECB9D

Verify Credential

Oracle Cloud Infrastructure 2025 Certified Data Science Professional

Oracle Certification 2025

Verify Credential

Get In Touch

I'm currently open to new opportunities and collaborations. Feel free to reach out if you'd like to discuss data science, AI projects, or just want to connect!