Technical Arsenal
Tools & Technologies
The stack I use to explore data and engineer solutions.
Bhargav Kumar Nath.
Data Scientist & ML Engineer.
Available for full-time work
Building predictive models on millions of real-world records, seamlessly bridging the gap between raw data pipelines and deployed ML systems.
bhargav@ml-cluster ~
$system-status --live
>
ALL SYSTEMS NOMINAL|GPU: 94% util|Uptime: 847d
4.5×
Conversion Uplift
£0.139
Forecasting MAE
<1ms
Inference Latency
8–32×
Serving Throughput
40%
VRAM Reduction
100M+
Event Pipeline
About Me
Researcher who ships
production systems.
I hold an MSc in Data Science and Analytics from the University of Leeds and a BTech in Computer Science and Engineering from Assam Don Bosco University. My expertise bridges Machine Learning and Software Engineering: not just building models, but deploying and scaling them reliably.
I specialize in predictive modeling, deep learning architectures, and end-to-end MLOps pipelines. From processing datasets exceeding 100 million records using PySpark to deploying high-performance inference engines, I focus on unlocking tangible value through robust data ecosystems.
Current Focus Areas
Building scalable prediction systems
Natural Language Processing & LLMs
Real-time analytics & streaming pipelines
Bridging data science research and production engineering
MSc
Data Science & Analytics, Leeds
~2
Years professional experience
11
End-to-end ML projects shipped
Philosophy
"A model is a mathematical fantasy, but an ML system is a living entity. I design for the shifting reality of the human world, not the static perfection of a laboratory."
01
Escape the State-of-the-Art Trap
Leaderboard victories rarely survive reality. I start with the simplest model to establish an honest baseline and prove if building ML is even necessary.
02
Algorithms Fade, Data is Foundational
Architectures change but long-term success depends on data quality. Real-world data is noisy and evolving; inflexible systems quickly become obsolete.
03
Deployment is the Starting Line
Standard software fails loudly, but ML systems fail silently via confident incorrect predictions. Production models need continuous monitoring to stay reliable.
Proof of Work
Data Science & ML Projects
Demonstrating end-to-end expertise uniting predictive modeling with resilient MLOps.
Engineering Track
Experience & Education
A track record of engineering impact across institutions.
Professional Engineering
M/S Sanjog TradingGuwahati, India
Jul 2020 to Nov 2021ETL Pipeline Engineering
Built Python ETL pipelines using pandas, Polars, and NumPy to process 800K+ sales records, automating data ingestion, transformation, and preparation for analytics workflows.
Time-Series Forecasting
Developed and evaluated time-series forecasting models using scikit-learn, XGBoost, statsmodels, and PyTorch, comparing performance with RMSE and custom metrics to support inventory planning.
Anomaly Detection & BI
Implemented an anomaly detection pipeline using Isolation Forest and deployed Streamlit and PowerBi dashboards to visualize forecasts, sales trends, and irregular transaction patterns.
IIT GuwahatiGuwahati, India
Jul 2022 to Aug 2022Database Architecture
Designed a normalized MySQL database schema with optimized indexing to manage user, scheduling, and transactional data across 15+ application modules, ensuring efficient data storage and retrieval.
Automated Logic & Analysis
Developed automated scheduling logic in Node.js to handle complex date constraints including holidays, leave periods, and scheduling conflicts, while analyzing user research data to guide application improvements.
Data Pipeline Formulation
Formulated data pipelines to aggregate and transform user interaction logs, facilitating the development of predictive models and enabling data-driven feature enhancements based on usage behavior.
Airports Authority of IndiaRegional HQ, India
Jul 2023 to Aug 2023Asset Data Analytics
Analyzed 1,053+ IT asset records using enterprise dashboards to track hardware lifecycle, deployment patterns, and maintenance data, identifying trends in asset utilization and replacement requirements.
ERP Workflow Evaluation
Evaluated procurement and vendor transaction data from the Government e-Marketplace (GeM) and documented SAP ERP data workflows across HR, Finance, and Procurement systems to understand data relationships.
Predictive Maintenance Groundwork
Automated the extraction and preprocessing of asset data using Python and Pandas, laying the groundwork for a predictive maintenance model that forecasts hardware failure rates based on historical lifecycles.
Academic Foundations
Sep 2024 to Nov 2025
MSc Data Science and Analytics
University of Leeds, UK
Specialized in advanced machine learning, predictive modeling, data mining, and big data architecture. Developed expertise in end-to-end data pipelines, real-time analytics, and MLOps principles.
Aug 2020 to Jun 2024
BTech Computer Science and Engineering
Assam Don Bosco University, India
Solid foundation in software engineering, algorithms, data structures, and database management. Led projects integrating classical software design with early predictive modeling applications.
Key Capabilities
Agentic AI Workflows
MLOps & ResOps
High-Performance Computing
Quantitative Research
End-to-End ML Pipelines
LLM Inference Optimization
Writing
Published Articles
Exploring algorithms, trends, and the intersection of technology, AI research, and society.