Dr. Ahilan kanagasundaram
SUMMARY
Accomplished machine learning expert with experience in proposing novel solutions and developing state-of-the-art systems for different applications.
Proficient in predictive modelling, data processing, data mining, and scripting languages, including Python.
Holding Australian and Sri Lankan Citizenship.
TECH STACK
● Python ● Linux environments ● Keras ● Deep Learning
● Keras ● Scikit-learn ● SQL ● Pandas
● PySpark ● Cloud ● PyTorch ● Signal Processing
EDUCATION
Queensland University of Technology (Apr 2010 – Nov 2014)
Australia
Ph.D (Machine Learning & Speech Processing)
University of Peradeniya (Feb 2004 – May 2008)
Sri Lanka
B.Sc. Engineering Honors - Electrical & Electronic Engineering
GPA: 3.80/ 4.00
WORK EXPERIENCE
Senior Lecturer & Machine Learning Specialist (Sep 2016 - Present)
University of Jaffna
Developed time series and deep learning models to accurately forecasting the load demand and solar PV power generation
Worked with other developers in a team to develop speaker verification and speaker diarization technologies for commercial applications
Developed machine learning based autism spectrum disorder detection system
Developed state-of-the-art deep learning-based speaker profiling systems
Proposed innovative approach to identify whether customer can pay the loan or not using speech utterance
Developed machine learning based automatic wind turbine maintenance scheduling system
Machine Learning Researcher (Nov 2014 – Sep 2016)
Queensland University of Technology
Proposed machine learning algorithms to improve the performance of speaker verification system in dataset mismatch and noisy conditions
Evaluated algorithms on terabytes of training data using parallelized routines and Sun Grid Engine (SGE)
Coordinated research projects within the SAIVT research program
Ph.D internship (Nov 2012 - Apr 2013)
Autonomous University of Madrid
Analyzed the insights of short utterance speech data using visualization and statistical approaches
Proposed novel machine learning approaches to compensate the unwanted variations
CERTIFICATIONS
Deep Learning Specialization – Coursera
Sequence Models – Coursera
Convolutional Neural Networks – Coursera
Improving Deep Neural Networks: Hyperparameter Tuning, Regularization and Optimization – Coursera
Neural Networks and Deep Learning – Coursera
Structuring Machine Learning Projects - Coursera
KEY ACCOMPLISHMENTS
Participated Speakers in The Wild (SITW) Challenge (2nd place in raking in the main core-core track)
Participated DIHARD speaker diarization challenge I and II
AHEAD research, innovation and commercialization grant
2 local patents and 1 international patent