Manish Kumar Roy

Data Engineer

Kolkata, West Bengal, India2 yrs 9 mos experience
AI EnabledAI ML Practitioner

Key Highlights

  • Achieved 99.5% precision in nut detection system.
  • Developed heart sound classification model with 92% F1 score.
  • Created personalized news recommendation system improving engagement by 20%.
Stackforce AI infers this person is a Data Science and Machine Learning professional with expertise in AI applications across various industries.

Contact

Skills

Core Skills

Artificial Intelligence (ai)Computer VisionDeep LearningSound Data AnalysisPredictive AnalyticsNatural Language Processing (nlp)Data AnalysisMentoringLeadershipImage ProcessingData VisualizationSoftware Development

Other Skills

AeronauticsAlgorithmsAnalytical SkillsArtificial Neural NetworksC (Programming Language)C++Collaborative Problem SolvingCommunicationData AnalyticsData AugmentationData CleaningData ScienceEducationEnglishExploratory Data Analysis

About

I am a passionate and results-driven Data Science and Machine Learning professional with a strong academic foundation in Aerospace Engineering from the Indian Institute of Technology (IIT) Kharagpur. My expertise spans Machine Learning, Deep Learning, Data Analytics, and Artificial Intelligence, with hands-on experience in developing and deploying advanced models to solve complex, real-world problems. I thrive on leveraging data-driven insights to create innovative solutions that deliver measurable impact. With proficiency in Python, Pandas, TensorFlow, and PyTorch, I have successfully implemented cutting-edge technologies such as Large Language Models (LLMs), Transformer-based architectures (BERT), and Deep Neural Networks (LSTM, ResNet, VGG). My work spans diverse domains, including Natural Language Processing (NLP), Computer Vision, and Predictive Analytics. For instance, I developed a personalized news recommendation system using Twitter activity analysis, which improved user engagement by 20% and increased click-through rates by 15%. I am deeply involved in research and innovation, having worked on projects like real-time electro-thermo-mechanical performance analysis of laser-welded joints, where I achieved an R² score of 0.98 using Random Forest. Additionally, I developed a heart sound classification model using Bidirectional LSTM, achieving an F1 score of 78%. My ability to apply advanced techniques like Polynomial Regression, Decision Trees, Random Forests, and XGBoost has enabled me to optimize models and deliver high-accuracy results. Beyond technical skills, I am a collaborative team player with experience as a Teaching Assistant and Student Mentor. I have guided students in aircraft design, optimization, and academic decision-making, fostering a positive and productive learning environment. My strong communication skills and ability to mentor others have been instrumental in driving success in both academic and professional settings. Currently pursuing a dual degree in Aerospace Engineering at IIT Kharagpur, I am eager to contribute to cutting-edge AI research and development. My goal is to leverage my skills in Data Science, Machine Learning, and AI to drive innovation and make a meaningful impact in the tech industry. Let’s connect! I’m always open to discussing new opportunities, collaborations, or innovative projects in AI, Machine Learning, and Data Science. Feel free to reach out—I’d love to hear from you!

Experience

2 yrs 9 mos
Total Experience
8 mos
Average Tenure
9 mos
Current Experience

Daitrix

Data Scientist

Sep 2025Present · 9 mos · Noida, Uttar Pradesh, India · Remote

Renata ai

AI and Application Software Engineer

Jun 2025Aug 2025 · 2 mos · Gurugram, Haryana, India · On-site

  • Developed industrial-grade real-time nut detection system using 2,254 images, identifying the presence/absence of 4 nuts
  • Implemented data cleaning, brightness normalization, CLAHE, data augmentation, and adaptive confidence thresholding
  • Deployed YOLOv8s achieving 99.5% mAP@0.5, 99.8% precision, 99.5% recall with multi-scale detection & fallback methods
  • Delivered a 20% manpower reduction, enhanced production accuracy by 17%, and accelerated inspection speed by 99.4%
Analytical SkillsArtificial Intelligence (AI)yoloOpenCVComputer Vision

Indian institute of technology, kharagpur

2 roles

Teaching Assistant

Jan 2025Apr 2025 · 3 mos · Kharagpur, West Bengal, India · On-site

  • Teaching Assistant for Aircraft Design and Optimization (AE49003) under Prof. N.K. Peyada and Prof. Anup Ghosh.
  • Evaluating and mentoring 4th-year students in aircraft design and optimization.
  • Providing feedback to improve students' design and optimization skills.
  • Supporting innovative solutions and aerospace engineering understanding.
  • Assisting in enhancing course materials for better learning outcomes.

Master Thesis Project | Guide:Prof Amardeep Ghosh

Aug 2024May 2025 · 9 mos · Kharagpur, West Bengal, India · On-site

  • Developed a Deep Learning model for heart sound classification using 832 audio files from an iPhone app and stethoscope
  • Utilized preprocessing, including volume normalization, noise reduction, audio segmentation, & MFCC for feature extraction
  • Engineered sophisticated Bidirectional LSTM model, optimizing layers and hyperparameters for sequential audio analysis
  • Attained an F1-Score of 92%, indicating effective precision & recall across the category, showcasing the models’ robustness
NumPyLong Short-term Memory (LSTM)KerasScikit-LearnSound dataEnglish+6

Tiptop technologies

Data Science Intern

May 2024Aug 2024 · 3 mos · Sunnyvale, California, United States

  • Project title: Automated News Recommendation Model Using Twitter Activity Analysis
  • Developed a personalized news recommendation system suggesting relevant articles based on users' Twitter activity and reading history
  • Applied collaborative filtering to analyze Twitter users' reading history and content-based filtering to match article content with user tweet
  • Utilized transformer-based BERT model embeddings to perform sentiment analysis, K-Means clustering, optimizing recommendation
  • Improved user engagement by 20% and significantly increased click-through rates by 15%, based on data from 10,000 user interaction
NumPyPredictive ModelingKerasPredictive AnalyticsScikit-LearnLarge Language Models (LLM)+11

Wmg, university of warwick

Research Intern

Apr 2024Jul 2024 · 3 mos · Coventry, England, United Kingdom · On-site

  • Project title: Real-time Electro-Thermo-Mechanical Performance of Laser Welded Joints in Li-ion Cells
  • Conducted real-time experiment collecting and preprocessing on approx. 30k data points on linear, circular, and rectangular laser weld
  • Applied Polynomial Regression, Decision Tree, Random Forest, XGBoost, and LSTM with hyperparameter tuning for optimization
  • Achieved R² score of 0.98 using Random Forest, Long Short Term Memory yielded R² scores of 0.85 for temperature and 0.74 for load data
NumPyKerasPredictive AnalyticsScikit-LearnEnglishResearch Skills+4

Upskill campus

Data Science and Machine Learning Intern

Jun 2023Jul 2023 · 1 mo · Noida, Uttar Pradesh, India

  • Project title: Transfer Learning Techniques for High-Accuracy Plant Disease Detection and Classification
  • Analyzed a comprehensive dataset featuring 20,500+ images across 15 plant species labeled with various diseases and health statuses
  • Applied data augmentation like resizing, cropping, and normalization, to significantly enhance the robustness of deep learning models
  • Utilized multiple deep learning architectures like ResNet50, VGG16, VGG19, and InceptionV3, achieving an accuracy of 98.5% with VGG19
  • Achieved a precision of 97.5% and a recall of 98.1% on the test dataset with VGG16, improving accuracy compared to traditional methods
NumPyImage ProcessingKerasScikit-LearnTransfer LearningData Science+8

Accenture

Data Analyst

Jun 2023Jul 2023 · 1 mo · Remote

  • Completed a simulation focused on advising a hypothetical social media client as a Data Analyst at Accenture
  • Cleaned, modelled, and analyzed 7 datasets to uncover insights into the content trends to inform strategic decisions
  • Prepared a PowerPoint deck and video presentation to communicate key insights for the client and internal stakeholders
PresentationsData AnalysisData CleaningEnglishMicrosoft ExcelData Analytics+4

Student welfare group, iit kharagpur

Student Mentor

Dec 2022Dec 2023 · 1 yr · IIT Kharagpur · On-site

  • Guided 4 students, ensuring a smooth transition to institute life and addressing diverse academic and extracurricular challenges with support
  • Conducted meetings, advising first-year students on academic and extracurricular decisions as part of the Dean's initiative at IIT Kharagpur
LeadershipEnglishOrganization SkillsCommunicationLearningMentoring

Indian institute of technology, kharagpur

Computer Vision Intern | Guide: Prof. Vikranth Racherla

May 2022Jul 2022 · 2 mos · Kharagpur, West Bengal, India · On-site

  • Developed an automated student attendance management system using face recognition technology
  • Implemented face_recognition library to detect and recognize faces in images and videos accurately
  • Integrated requests library for seamless communication, sending output images and CSV files to Telegram
PythonKerasOpenCVComputer Visionface_recognition LibrarySoftware Development

Education

Indian Institute of Technology, Kharagpur

M.tech — Aerospace Engineering

Jul 2024Jul 2025

Indian Institute of Technology, Kharagpur

Aerospace engineering — B.tech

Dec 2020Jul 2025

Resonance Eduventures Limited

Jun 2019Jun 2020

Sudhir Memorial Institute

Higher secondary

May 2017Mar 2019

Sun Shine School, Rosera

secondary

Mar 2005Mar 2017

Stackforce found 100+ more professionals with Artificial Intelligence (ai) & Computer Vision

Explore similar profiles based on matching skills and experience