Production ML Systems
Building scalable machine learning pipelines from research to production, with expertise in model deployment and MLOps workflows.
Dhrubajyoti (DJ) Ray
AI-Focused Data Scientist
Building scalable ML systems that drive real-world impact. 2+ years deploying production-grade models in semiconductor manufacturing, with expertise in deep learning, LLMs, and MLOps.
📧
dtray@tepper.cmu.edu
📱
+1 (412) 979-6327
0
% Reduction in Wafer Measurement Time
0
$M Annualized Yield Recovery
0
% Operational Throughput Boost
Technical Expertise & Specializations
Specialized in building production-grade ML systems with deep domain knowledge in semiconductor manufacturing
Advanced Analytics
Deep statistical modeling and feature engineering expertise, with robust ETL workflows for complex data processing at scale.
AI & LLMs
Cutting-edge AI development with transformer architectures, large language models, and retrieval-augmented generation systems.
Technical Proficiency
Python
95%
SQL
90%
TensorFlow
88%
PyTorch
85%
AWS
82%
GCP
80%
Docker
85%
Kubernetes
75%
Semiconductor Manufacturing
Specialized domain expertise in semiconductor fab processes, yield optimization, and quality control systems
Notable Projects & Impact
Showcasing technical innovation and measurable results
Production ML at Micron Technology
30% Time Reduction
Machine Learning
Python
Regression Models
Real-time Dashboards
Engineered advanced regression models that revolutionized wafer measurement processes, while implementing AI-driven defect detection systems that significantly improved manufacturing quality control.
• 30% faster wafer measurement
• 15% improved fault detection
• 2 weeks faster diagnostics
Dark Souls LoreGraph System
End-to-End AI
LLMs
Neo4j
Streamlit
Cypher
NLP
Built a comprehensive AI pipeline leveraging Large Language Models for entity extraction, integrated with Neo4j graph database and an intelligent QA system featuring natural language to Cypher translation.
• Complete AI pipeline
• Graph database integration
• Natural language queries
View Dark Souls Project
AI-Powered Cold Email Agent
15% Response Rate ↑
GPT-4
RAG
FastAPI
LLM Agents
Developed an autonomous LLM agent powered by GPT-4 and Retrieval-Augmented Generation, creating personalized cold email campaigns that significantly improved engagement rates.
• 15% higher response rates
• Autonomous email generation
• FastAPI integration
CFPB Complaints Analysis
30% Faster Resolution
NLP
UMAP
KMeans
Streamlit
Big Data
Processed and analyzed 1.6 million consumer complaints using advanced NLP clustering techniques, redesigning complaint taxonomy and creating interactive dashboards for faster issue resolution.
• 1.6M complaints processed
• 30% faster issue resolution
• Interactive dashboards
Professional Journey
MT
Data Scientist 2
Micron Technology
Jul 2022 – Jul 2024 | Hyderabad
- Achieved 30% reduction in wafer measurement time through advanced analytics
- Improved defect detection accuracy by 15% using machine learning models
- Delivered $10M annualized yield recovery through production deployment optimization
- Boosted operational throughput by 20% via manufacturing integration
BD
Product Analyst
ByteDance
Mar 2021 – Aug 2021 | Mumbai
- Built churn prediction models increasing user retention by 5%
- Developed comprehensive Tableau dashboards for A/B testing analytics
- Automated error reporting systems reducing QA cycles by 30%
Education
CMU
Carnegie Mellon University
Master of Science in Business Analytics
GPA: 4.0/4.0 (Current)
IIT
IIT Kharagpur
Integrated B.Tech + M.Tech Chemical Engineering
GPA: 8.2/10 (2017-2022)
Ready to Build Something Amazing Together?
Looking for an AI-focused Data Scientist who can bridge the gap between research and production? Let's discuss how my experience in scalable ML systems and manufacturing optimization can drive impact for your team.
Currently pursuing MSBA at CMU, available for full-time opportunities starting May 2025
Pittsburgh, PA | Open to relocation