Naveen Johnson V - Portfolio

About Me

Data Scientist with MSc in Data Science (EURECOM) and 2+ years enterprise analytics experience. Proficient in Python, PyTorch, TensorFlow, NLP, deep learning, Generative AI, and knowledge engineering. Seeking Junior Data Scientist roles; open to Junior Data Engineer/Analyst positions.

Technical Skills

Programming Languages

Python R SQL JavaScript

Machine Learning & AI

Scikit-learn TensorFlow PyTorch HuggingFace LangChain

Data Analysis

Pandas NumPy Matplotlib Seaborn

NLP & Deep Learning

BERT Transformers SpaCy NLTK

Cloud & DevOps

AWS GCP Docker Git CI/CD

Specializations

Computer Vision NLP Knowledge Graphs ETL

Key Competencies

Data Cleaning & Preprocessing
Exploratory Data Analysis
Statistical Modeling
Feature Engineering
Model Evaluation & Optimization
Data Visualization
ETL Process Design
Business Intelligence

Professional Background

Download Full Resume

Professional Experience

Data Science Research Intern - AI Planning / NLP

ALTEN Labs March 2025 - August 2025 | Valbonne, Alpes-Maritimes, France

Built a template-driven LLM model pipeline, using OpenAI's GPT-4o, that generates formal PDDL task specifications, which are then executed with Fast Downward open-source planner to create playable interactive narratives
Added schema/consistency checks with automated fixes, cutting execution failures and improving reliability by ~60%
Implemented state hashing and content-addressed caching; built utilities for plan replay and auditability

Associate Consultant - SAP Analytics

Blueprint Technologies Pvt Ltd December 2021 - August 2023 | Bengaluru, Karnataka, India

Delivered 20+ SAC dashboards with drill-downs and detailed statistical analyses for pharma and agrochem stakeholders
Built ETL migrations between SAP ECC and BW/4HANA (ADSOs, CompositeProviders, transformations)
Designed KPI logic and led data reconciliation, and UAT for production reports; delivery across 5+ enterprise projects

Intern - SAP Analytics

Blueprint Technologies Pvt Ltd July 2021 - December 2021 | Bengaluru, Karnataka, India

Completed 200+ hours across SAC, BW, HANA, Python, SQL; contributed prototypes and ETL/UAT documentation
Automated data-quality checks using Python/Pandas and SQL extracts

Research Projects

LLMs for Automated Generation of Interactive Scenarios

AI Planning | NLP | PDDL | LLM

Developed a neuro-symbolic pipeline combining LLMs with classical planners to generate playable interactive narrative scenarios. Implemented planner-validated pipeline with ~60% reliability improvement through automated schema/consistency checking and state caching for auditability.

Python GPT-4o PDDL Fast Downward

Knowledge Engineering in the LLM Era

Knowledge Graphs | NLP | RDF | Ontologies

Reproduced and extended Text2KGBench (Text-to-Knowledge-Graph Benchmark) experiments comparing state-of-the-art LLM models for RDF triple extraction, evaluating precision, recall, and F1-score. Built ontology-aware LLM pipeline achieving superior performance with GPT-4o (AUC 0.89) over Qwen2.5 32B. Integrated DBpedia-WebNLG, Wikidata-TekGen, Odeuropa datasets with CIDOC-CRM-aligned ontologies to evaluate semantic accuracy and hallucination reduction.

Python PyTorch HuggingFace RDF

View Project

Sentiment Analysis using BERT

NLP | BERT | Transformers

Implemented BERT-based model for tweet sentiment classification achieving 78% F1-score. Compared performance with traditional NLP approaches (TextBlob, VADER), demonstrating superior results with transformer models through comprehensive preprocessing pipeline.

Python BERT NLP PyTorch

View Project

Audio Anomaly Detection in Industrial Machinery

Deep Learning | Autoencoders | PyTorch

Implemented Convolutional and Variational Autoencoders for detecting anomalies in industrial machine sounds. Achieved 88.8% AUC score with VAE outperforming classical AE (76.6%), enabling predictive maintenance applications for slide rail machinery.

PyTorch VAE Signal Processing

View Project

Aerial Cactus Identification

Deep Learning | Computer Vision | CNN

Developed custom CNN architecture for columnar cactus identification from aerial imagery using stacked convolutional blocks with batch normalization and dropout. Achieved 99% accuracy and ROC/AUC score of 1.0 on VIGIA project dataset.

Python TensorFlow Computer Vision Keras

View Project

Solar Power Generation Forecasting

Machine Learning | Time Series | Regression

Comparative analysis of LASSO regression and k-Nearest Neighbors models for day-ahead DC power prediction from solar plant. LassoCV achieved RMSE of 0.538 using lagged DC power and ambient temperature features on hourly data.

Python Scikit-learn Time Series

View Project

Education

Master of Science in Computer Science, Data Science Specialization

EURECOM, Biot, France 2023 - 2025

Core Subjects:

Machine Learning
Reinforcement Learning
Advanced Statistical Inference
Deep Learning
Distributed Systems and Cloud Computing
Database Management System Implementation (DBSys)
Digital Image Processing & Video Compression
Semantic Web and Web Interaction Design
System and Network Security
Secure Communications
Project Management and Research Methodology

Diploma in Data Science

Indian Institute of Technology, Madras, Chennai, India 2021 - 2024

Comprehensive program covering statistics, mathematics, data science fundamentals, Python programming, and SQL.

Bachelor of Technology in Electronics and Communication Engineering

National Institute of Technology, Calicut, Kozhikode, Kerala, India 2016 - 2020

Strong foundation in engineering fundamentals with focus on signal processing and communication systems.

Certifications & Test Scores

Test Scores

TFI (Test de français international)

French Language Test

January 2025

Score: 385 - Demonstrates B1 level proficiency

TOEFL

English Language Proficiency Test

November 2022

Score: 110 - Demonstrates C2 level proficiency

Professional Certifications

Introduction to SAP Datasphere

SAP

Issued: June 2023

Skills: Data Visualization, Data Modeling, SQL

Foundational Level in Programming and Data Science

Indian Institute of Technology, Madras

Issued: August 2022

Skills: Statistics, Mathematics, Data Science, Python, SQL

Full-Stack Engineer Career Path

Codecademy

Issued: March 2021

Skills: Git, Version Control

Cybersecurity and the Internet of Things

Coursera

Issued: December 2020

Credential ID: NFTWGKHJQXEP

Python Data Structures

Coursera

Issued: December 2020

Credential ID: 9WCQMTPJPJ4Q
Skills: Python Programming

Programming for Everybody (Getting Started with Python)

Coursera

Issued: November 2020

Credential ID: RF4AUX2KPXDF
Skills: Python, Git

Elements of AI

University of Helsinki and Reaktor

Issued: June 2018

Credential ID: j1nbixsojic
Skills: AI Fundamentals

Get In Touch

Are you looking for a data-driven professional to help solve complex problems or drive insights from your data? Or just want to say "Hi 👋"? Feel free to reach out!