PRANAVA KAILASH
SUBRAMANIAM PREMA

Building the pipelines and models that turn raw data into decisions.

Who I Am

I'm a Data Engineer and Machine Learning specialist with an M.Sc in Data Science from the University of Surrey. I build end-to-end data systems — from ingestion and transformation to model training and deployment.

My work spans Python, SQL, and modern data platforms including AWS, Databricks, and Apache Airflow. I design ETL pipelines that are reliable, auditable, and built to scale.

I'm particularly interested in how data infrastructure, ML models, and AI agents intersect — systems that don't just analyse data, but act on it.

Data Engineering

ETL pipelines, orchestration, and cloud data platforms

Machine Learning & AI

Model training, NLP, deep learning, and LLM applications

AI Engineering

RAG systems, AI agents, and on-device AI with vector databases

Pranava Kailash

M.Sc Data Science · University of Surrey · 2025

Guildford, United Kingdom

My Journey

Feb 2024 – Feb 2025

M.Sc Data Science (Merit, 63.42%)

University of Surrey, UK

  • Built a Cybersecurity NER model using DeBERTa, achieving 91.88% F1-score, deployed as a FastAPI + SQLite API endpoint for CTI analysts
  • Optimized CNNs with metaheuristic algorithms (GA, HS, FA) for image classification on CIFAR-10, improving accuracy and convergence
  • Developed containerized web application with Flask and Docker; created predictive models for sentiment analysis and housing price prediction
  • Certified by the EDI/University of Surrey Future Leaders Programme (Oct–Nov 2024)

Oct 2022 – Jan 2024

Data Researcher / Engineer (Freelance)

Fleet Street Research, UK

  • Built Python data extraction pipelines — 50% gain in data acquisition efficiency
  • Integrated MySQL + SQLite pipelines via SQLAlchemy with zero data loss

Aug 2019 – May 2023

B.E Computer Science & Engineering (Distinction, 9.5 CGPA)

SNS College of Technology, India

  • Published research paper on Plant Disease Detection using ML
  • Founded and ran "REGEX - CSE" club for peer learning and networking
  • Won multiple Ideathon events and the "Centre for Creativity" Award

Feb 2022 – Apr 2022

Data Scientist Intern

Yoshops.com, India

  • 90% accurate deep learning model (VGG16) for osteoarthritis detection
  • Automated web scraping — 40% improvement in real-time pricing accuracy

Mar 2021 – Jun 2021

Data Scientist Intern

Forsk Coding School, India

  • NLP sentiment analysis on customer reviews using transformer-based models
  • Deployed scalable ML models on AWS via Flask API

Certifications

AWS Databricks Platform Architect·Databricks·Apr 2025Docker Foundations Professional·Docker Inc·May 2025Career Essentials in GitHub·GitHub·May 2025SQL Intermediate·HackerRank·Aug 2025Power BI Beginner to Pro·Pragmatic Works·Mar 2025

Featured Projects

A collection of my work across machine learning, data engineering, and AI applications

Latest ReleaseAI Engineering

Prompt Enhancer

Chrome extension using Google's Gemini Nano AI for on-device prompt enhancement. Enhances ChatGPT prompts instantly with 100% local processing, zero API costs, and complete privacy.

JavaScriptChrome APIGemini NanoAI
FeaturedMachine Learning

CyNER_2.0_API

Advanced Named Entity Recognition (NER) system designed to identify cybersecurity entities with 94% accuracy. Features a REST API for real-time NER inference on cybersecurity text data.

PythonNLPMachine LearningFlaskspaCy

Other Projects

Technical Skills

A comprehensive toolkit spanning data science, machine learning, and software development

Programming Languages

PythonC++JavaScriptSQL

Machine Learning & AI

PyTorchTensorFlowscikit-learnNLPDeep LearningLLMsNumPyPandas

Data Engineering

Apache AirflowETL PipelinesMySQLSQLiteFlaskFastAPIMongoDB

Cloud & DevOps

AWSGoogle CloudDockerCI/CDDatabricks

Data Visualization

PlotlyPower BIMatplotlibSeaborn

Currently Expanding

Cloud Certifications

AWS & GCP certifications

MLOps & Deployment

CI/CD for ML models, experiment tracking

LLM & Agent Systems

Building AI agents, RAG pipelines, tool-augmented LLMs

Real-time Systems

Streaming data & live dashboards

Blog & Insights

Writing about AI, data, and how these tools actually work behind the scenes—now live on Medium.

Latest on Medium

Prompt Enhancer Chrome Extension with Gemini Nano

Demystifying AI Tools: How They Work Under the Hood

Over the last few weeks, I've been building a Chrome extension called Prompt Enhancer that turns rough ideas into clear, structured prompts for ChatGPT in a single click. It runs fully on-device using Chrome's built-in Gemini Nano via the Prompt API, so nothing ever leaves your browser.

Follow @pranavakailash on Medium for new posts.

Let's Connect

I'm open to roles in data engineering, machine learning, and AI — based in the UK and available now. If you're building something that needs strong data infrastructure or applied ML, I'd like to talk.