HomeSkillsExperienceWorkContact
Available for hire

Sankar Dev S

Full-stack Engineer & AI Researcher.
Building the bridge between Conceptual AI and Production Reality.

London, UK

Core Technologies

AI/ML
Next.js
Python
TypeScript
React

Technical Arsenal

A comprehensive toolkit for building scalable AI solutions and robust web applications.

AI & Machine Learning

NLP (BERT, RoBERTa)Deep Learning (TensorFlow, CNN)Generative AI (Prompt Engineering, RAG, Langchain)Classical ML (Logistic Regression, Random Forest)PySparkLLM Data PipelinesSentiment Analysis

Web Development

JavaScript/TypeScript (React.js, Node.js, Express.js, Next.js)Python (Django, Flask)PHPHTML5CSS3

Cloud & DevOps

AWS (EC2)DockerCI/CDGitRESTful APIsSockets.IOPostmanSQL

Professional Skills

Project ManagementTechnical LeadershipBespoke Software ArchitectureTechnical DocumentationCode ReviewsPerformance Optimization

The Journey

Professional milestones and academic achievements.

Research Assistant

University of East London
September 2024 – Present
London, United Kingdom
  • Engineered a high-performance data pipeline to merge and preprocess 250,000+ text samples from multi-domain datasets.
  • Optimized training latency by 30% by implementing custom cleaning scripts and efficient tokenization strategies.
  • Developed the core classification architecture for JDSIS-published research, achieving a peak 95.08% F1-score in text-based emotion recognition using RoBERTa.
  • Pioneered a multimodal framework (SoundSense) using Wav2Vec2.0, outperforming traditional SVM/RF models by 30% in vocal anomaly detection accuracy.

Software Associate

Riss Technologies
May 2022 – May 2023
Ernakulam, Kerala, India
  • Architected and delivered 200+ bespoke software solutions, maintaining a 95% client satisfaction rate.
  • Integrated predictive ML models into production environments to automate decision-making processes, reducing manual data processing time by approximately 20%.
  • Developed and deployed responsive Python (Django/Flask) and PHP web applications, focusing on scalable backend architecture and high-performance user interfaces.

Freelance Web Developer

Techise Solutions
February 2023 – October 2023
Alappuzha, Kerala, India
  • Architected scalable backend systems using Laravel and MySQL, optimizing database schemas.
  • Engineered custom front-end components using React.js and modern CSS frameworks, ensuring 100% mobile-responsive designs.
  • Managed end-to-end project lifecycles, from initial requirement gathering to final AWS/VPS deployment, ensuring consistent on-time delivery.

MERN Stack Internship

Camerinfolks
September 2022 – January 2023
Ernakulam, Kerala, India
  • Acquired foundational knowledge of React JS and Node JS during an intensive internship, gaining proficiency in core concepts and principles.
  • Strengthened skills through hands-on experience, significantly impacting to the development of robust and scalable web applications.
  • Demonstrated adaptability and quick learning, enhancing the success of projects while continually expanding expertise in these technologies.
  • Implemented and championed best practices, optimizing development workflows for increased efficiency.

Education

Masters in Artificial Intelligence

Distinction

University of East London, UK

2023 – 2025

Bachelor of Computer Application

Kerala University, India

2019 – 2022

Selected Work

Research papers and full-stack applications.

SoundSense – Multimodal Vocal Anomaly Detection

Developed a cross-modal AI system to detect emotional inconsistencies between spoken text and vocal tone using Wav2Vec2.0 and RoBERTa.

Wav2Vec2.0RoBERTaPython

EmoTract – Advanced NLP Framework with Age Verification

Engineered a real-time sentiment platform using BERT and RoBERTa to classify 28 distinct emotions. Architected the full stack using Django, React.js, and TypeScript.

BERTRoBERTaDjangoReact.jsTypeScriptDocker

Latest Publication

"NLP-Framework for Youngsters with Advanced Transformer-Based Models"

Journal of Data Science and Intelligent Systems (JDSIS), 2026

  • Developed a sentiment analysis framework achieving a 95.08% F1-score using fine-tuned RoBERTa models.
  • Benchmarked BERT & RoBERTa against traditional models (LR, RF) to validate superior performance incomplex emotion classification.

Let's Connect

I'm currently open to new opportunities in AI Research and Full-Stack Engineering. Whether you have a question or just want to say hi, I'll try my best to get back to you!

s4nkar.connect@gmail.com
London, UK
+44 7460054747

© 2026 Sankar Dev S. All rights reserved.