Hi, I'm Dinuja Perera
Machine Learning Engineer
I am interested in working with machine learning and data science teams on real-world projects. My focus areas include deep learning, NLP with transformer-based models, and time-series analysis. I am particularly looking for opportunities where models are developed, deployed, and maintained in production environments.
Match my experience to your role
Paste your job description and I'll highlight the most relevant skills, projects, and research instantly.
Transforming Data into Intelligence
Machine Learning Engineer with a passion for building AI systems that solve real-world problems
AI Researcher & Engineer
With a proven track record in developing machine learning solutions, I help organisations harness data-driven insights to drive strategic decisions and innovation.
UK Global Talent
Endorsed by UK Research and Innovation (UKRI)
AI Specialist
Generative AI, LSTM, RNN, and NLP expert
Collaborative
Cross-functional team experience
Impact Driven
Real-world AI solutions for industry
Core Competencies
- • Generative AI solutions for enterprise
- • Predictive maintenance and time-series analytics
- • NLP pipelines and conversational AI
- • Cloud-native ML deployment strategies
Professional Journey
Building AI solutions across industries and academia
KTP Associate – Machine Learning Engineer (Predictive Maintenance)
Uniper UK | Loughborough University
- ●Built and tested generative AI models (LSTM, ESN, RNN) on turbine sensor time-series data to forecast signals and detect early signs of gas turbine trips
- ●Applied unsupervised learning (K-Means clustering) to group sensor behaviours and improve feature engineering for anomaly detection
- ●Implemented a statistical correlation-based model to identify faulty instruments, comparing autocorrelation patterns of healthy vs. faulty sensors
- ●Processed and analysed large-scale sensor datasets in Azure Databricks using GPU acceleration to speed up deep learning training
- ●Designed interactive dashboards to share diagnostic results and visual insights with engineers and managers
- ●Worked closely with government, academic, and industry partners to align technical work with operational needs
Machine Learning Engineer
CML Insight Inc., Texas, USA
- ●Cleaned, merged, and standardised multiple datasets to ensure smooth integration for downstream analysis
- ●Engineered NLP features (tokenisation, embeddings, sentiment scores) from text data to support model development
- ●Assessed feature importance and checked for potential target leakage to improve model reliability
- ●Developed cohort-based models to analyse behavioural patterns across different customer groups
- ●Supported the team by onboarding and mentoring interns on basic ML workflows and coding practices
- ●Maintained ML pipelines to enable model reuse on future incoming data
Teaching Assistant
Department of Information Systems Engineering, University of Colombo
- ●Conducted practical sessions and supervised projects for master's and undergraduate courses, including Machine Learning and Neural Computing, Data Analytics, and Embedded Systems
- ●Co-ordinated grading assignments, conducting code reviews, and invigilating exams for undergraduate and master's level programmes
Data Analyst
Brandix Apparel Limited
- ●Analysed spare parts datasets using Microsoft Excel pivot tables to ensure data accuracy
- ●Collaborated with cross-functional teams to identify and rectify data quality issues
- ●Conducted assessments of discrepancies and inaccuracies to enhance overall data integrity
Engineering Trainee
Airport & Aviation Services Sri Lanka
- ●Familiarised with navigation security surveillance communication systems
- ●Participated in preventive maintenance activities to ensure optimal performance
- ●Assigned to the Department of Air Navigation and Engineering to enhance operational efficiency
Tech Stack & Expertise
A comprehensive toolkit for building intelligent systems
Languages & Frameworks
Cloud & Infrastructure
Databases
Analytics & Visualisation
ML Models & Techniques
DevOps & Collaboration
Specialised In
Accelerated Computing
Forecasting & Analysis
Maintenance Systems
Featured Research & Projects
Showcasing research, production ML systems, and experimentation
Transformer-Based Sentiment Analysis of Company Reviews
MSc Dissertation - NLP & Interactive Web Application
Leveraged NLP and transformer-based models to analyze Glassdoor employee reviews from 500 UK companies. Compared BERT, DistilBERT, RoBERTa, DeBERTa, and XLNet, with XLNet achieving 76% accuracy. Integrated topic modeling (LDA, NMF) and aspect-based sentiment analysis (NER, POS tagging) to identify key themes. Built an interactive React-based web interface for company comparison and insight generation.
Employee Churn Risk Prediction and Behavioural Analytics
Causal Machine Learning for Workforce Analytics
Developed causal machine learning models to understand employee turnover behaviours. Used Random Forest and statistical feature importance techniques to identify the most influential drivers of churn. Performed feature leakage detection, refined model input space, and enhanced model generalisation. Integrated email sentiment analysis as an additional behavioural signal. Work carried out in a Linux-based environment using object-oriented Python.
Predicting Road Accident Severity in the UK
Supervised Machine Learning Classification
Applied supervised machine learning to predict the severity of road accidents across the UK using 2019 public data. Evaluated Random Forest, SVM, Decision Tree, KNN, and Deep Neural Networks. The deep neural network achieved the highest accuracy of 80.65% in classifying accidents as 'Slight,' 'Serious,' or 'Fatal.'
Metal Part Lifespan Prediction and Defect Classification
Regression & Defect Detection with ML
Investigated metal part manufacturing datasets to predict part lifespan and classify defects. Regression models (Linear, Lasso, Ridge, Random Forest) were compared, with Random Forest achieving 97% accuracy. For defect detection, both binary classifiers and CNNs were tested. K-Means clustering revealed distinct process parameter groups influencing part quality.
Comparative Analysis of BART and RoBERTa for Hate Speech Detection
Published Research - WiNLP 2022
Explored transformer-based approaches for detecting hate speech on YouTube and Reddit using the ETHOS dataset. Compared BART and RoBERTa for binary and multi-class classification. BART achieved 70% F1-score and 58% top-1 accuracy, outperforming RoBERTa in distinguishing hate categories including gender, race, and religion.
WiNLP Workshop co-located with EMNLP 2022
Benchmark NLP Algorithm for Hate Speech Detection
Deep Learning on Social Media
Tested 12 deep learning architectures, including RNNs, CNNs, transformer-based models (e.g., BERT, RoBERTa), and hybrid architectures (e.g., CNN + LSTM) to detect hate speech on social media platforms.
Published Research
"Short Comparative Analysis on Pretrained BART and RoBERTa in Detecting Hate Speech on YouTube and Reddit Platforms"
Presented at WiNLP Workshop co-located with EMNLP 2022
Academic Excellence
Building a strong foundation in data science and machine learning through rigorous academic training
Tier 1 Global Talent Visa
Endorsed by UK Research and Innovation (UKRI)
Published Research
Presented at WiNLP Workshop co-located with EMNLP 2022
Multiple Distinctions
Achieved distinction grades in MSc Data Science and Graduate Diploma
MSc Data Science
2023 – 2024University of Greenwich, London
Key Modules: Machine Learning, Applied Machine Learning, Data Visualisation, Statistical Methods for Time Series Analysis, Graph and Modern Databases, Big Data, Blockchain for FinTech Applications.
MSc specialising in Data Science, Analytics and Engineering
2021University of Moratuwa, Sri Lanka
Key Modules: Machine Learning, Pattern Recognition, Data Mining, Data Analytics, Advanced Databases, Business Intelligence, Neural Networks, Advanced Algorithms, Statistical Inference, Bioinformatics.
Grad. Dip. in Electronics Telecommunication & Computer Engineering
2020Institution of Engineers, Sri Lanka
Key modules including Digital Signal Processing, Computer Security, Computer Networks, and Communication Engineering.
BSc (Electronics & IT)
2015-2019University of Colombo, Sri Lanka
Key modules including Applied Mathematics, Statistics, Computer Science, and Physics.
Professional Certifications
Leadership & Management - CMI Level 5 Equivalent
Ashorne Hill Management College, UK
British Airways Data Science Job Simulation on Forage
April 2024
Advanced NLP
LinkedIn Learning
PySpark for Big Data
LinkedIn Learning
Azure DevOps Fundamentals
LinkedIn Learning
Generative AI Foundations
LinkedIn Learning
Let's Work Together
Whether you have to discuss opportunities, ask a question, or just want to say hi, feel free to reach out!
Location
Nottingham, UK
Ready to Connect?
I'm currently available for full-time possitions in Machine Learning and AI.
Looking for Collaboration?
I'm particularly interested in projects involving Applied Machine Learning, MLOps, Cloud Computing, NLP, and Time-Series Analysis.
Let's discuss how we can work together to build innovative AI solutions.