Recruiter LogIn

Don't have an account? Register now

Forgot Password?

Recruiter Registration

Custom

Select Your Default Landing Page

Photo

AMAR ****** Deep Learning Engineer

$600 / day

Summary

In my dynamic 2.5-year tenure as a Data Scientist at Veolia, I have specialised in analysing complex datasets to drive strategic business decisions. I am proficient in Python, SQL, Tableau, and advanced ML algorithms, excelling in analysing, visualising, and communicating insights. I aim to combine my data science skills with project management expertise to make impactful contributions.

SKills

3D Reconstruction and Registration (PCLand Calibration • Deep Learning and Neural Networks Architectures ◦ Convolutional Neural Networks (CNNAttentionBash and MATLAB)Machine Learning and Data analysis tools (TensorFlowBlender) • Nonlinear Optimization (C++ ceres and MANOPT for Optimization on Manifold) • Robotics platform (ROS and Gazebo) • Rigid and Non-Rigid SLAMC++ChatBotClassificationCloud CompareDecision TreesGRUKerasLanguages (PythonMachine Translation) • Supervised Machine Learning (RegressionMatplotlibMeshLabMobileNetNeural Style TransferNLP and Recurrent Neural Networks LSTMNumPyObject Detection and RecognitionOpen3DPandasPyMeshlabPyTorchRandom Forest and XGBoost) • UnsupResNetscikitlearnseaborn and scipy) • Computer Vision and Image Processing (OpenCV in Python and C++) • Point cloud ProcessingSemantic Segmentation) ◦ Sequence ModelsSpeech RecognitionSQLState Estimation and Pose Graph OptimizationTransfer LearningTransformersU-Net
AI ConsultantDeep Learning EngineerMachine Learning Engineer

Education

March 2019 - Dec 2020 Master of Data Science at The University of Melbourne
July 2014 - May 2018 Bachelor of Computer Science at VIT University

Experience

Oct 2021 - Present AI MODEL INDUSTRIALISATION at Veolia

– Implemented BurstID, an AI solution that optimizes pipe renewal plans to minimize water losses from networks, proving effective in Australia, New Zealand, and China.
– Developed GreenPath Online, Veolia’s solution for automatic monthly emissions tracking, enhancing awareness of carbon footprints in water treatment plants and supporting
targeted emission reduction plans for 2025.
– Innovated Virtual Submetering Trial, which divides energy consumption by equipment in facilities, reducing labor hours significantly compared to manual energy assessments
– Analysed data from water utilities to assist in customer profiling, identifying high-consuming customers, and helping them to run water-saving campaigns
– Accountable for code production, including constant maintenance of the code and re-factoring when necessary.

Mar 2020 - Dec 2020 Data Scientist at The University of Melbourne

– Identified key blockages in the University admissions data with their process time for international student course applications that take longer and have more transactions.
– Created efficient workflow combinations for the University to expedite the application process, ensuring minimal delays. Utilised dashboard visualisation to streamline
operations, resulting in a reduction from 8 to 7 weeks.
– Executed data exploration on a 1.5GB dataset to extract key features from 54 features in admission data. This involved indentifying correlations between various workflow
processes of appications and the corresponding completion times.
– Conducted daily, experiements involving data cleansing, processing and the application of Python-based Machine Learning classifiers. These efforts aimes to predict both the
likelihood of applications success post-offer release and the time taken by applicants to accept offers.