Data Scientist Machine Learning
Resume:
ASHWIN PANCHAPAKESAN
@inspectorG*dget
ad6zk5@r.postjobfree.com
INSPECTORG4DGET.COM
RELEVANT WORK AND RESEARCH EXPERIENCE
Data Scientist July 2020 – Present
SecDev Group, Ottawa, Ontario
• Patent-pending Language Agnostic Topic Modeler
o Computed word embeddings and performed clustering to extract topics from corpora in any language
• Automated and deployed the core OSINT data collection platform o Set up data collection methodology and tooling with APIs, web scraping and device emulation o Horizontally scaled components using multiprocessing and multiple nodes to improve execution time o Redesigned PostGreSQL schemas for improved extensibility and lookup performance
• Built a predictive dashboard to contextualize data analytics
• Built COVID-19 vaccination rollout simulator
o Simulated resident vaccination timelines based on city-scale policy decisions using MapBox, Streamlit o Built relationships with external experts to improve simulation integrity and veracity Data Scientist July 2019 – June 2020
ROKU Inc (ROKU), New York, NY
• Designed and optimized machine learning algorithms to maximize advertising impact o Pulled data from distributed data store using Hive o Designed and tested machine learning models
• Built relationships between my team and the Engineering team to facilitate smoother deployment
• Reduced development time by augmenting the team’s tooling Part-time Professor
University of Ottawa, Ottawa, Ontario
September 2017 – December 2017
• Taught Introduction to Computing I (python) to first year students
• Delivered lectures, held office hours, created evaluations, managed TAs Teaching Assistant
University of Ottawa, Ottawa, Ontario
September 2016 – August 2018
• Conducted labs, evaluated assignments, and held office hours
• Delivered course lecture in professor’s absence
• Course topics included Python, Java, engineering ethics, algorithms design Student Research Intern May 2014 – September 2014
Center for Operational Research and Analytics, Defence Research and Development Canada, Ottawa
• Used evolutionary algorithms to optimize mission parameters
• Used MATLAB and C to run simulations and optimize (on multiple metrics) device usage schedules
• Evaluated and optimized evolutionary optimization algorithms (Non-dominated Sorting Genetic Algorithm, etc)
Software Development Engineering Intern
Amazon.com, Seattle, WA
May 2012 – August 2012
• Used Genetic Algorithms to improve accuracy of display advertising prediction by 36x
• Co-inventor on a patent application relating to modeling attribution of advertisement features Radio Automation Software Testing Intern
Research In Motion, Ottawa, Ontario
May 2011 – September 2011
• Developed automated test scripts in Python to test wi-fi capabilities of BlackBerry handhelds and tablets
• Developed a framework to decrease further test generation time and latency
• Improved efficiency of test case analysis by over 20,000% Technical Skills
- Git
- GitHub Actions
- Amazon AWS
- SageMaker
- Google Compute
Platform
- Terraform,ansible
- Docker
- Sentry.io
- MDRaid
Programming
Languages
- Python3
- Bash
- Hive
- PostGreSQL
- LaTeX
Python Stacks
- SciKit Learn
- Numpy
- Pandas
Authored
Packages in PyPI
- Pyvolution
- Pystitia
Advanced
Concepts
- Evolutionary
algorithms
- Combinatorial
algorithms
Fluent Natural
Languages
- English
- Tamil
Extra Curricular
- Teaching STEM
to underprivileged
children
Workspace
administrator
- Developed
FinTech platform
for an India
microfinance
company
- Skiing
- Cooking
EDUCATION
Ph. D. Computer Science
University of Ottawa, Ottawa, Ontario
September 2013 – June 2019
Specialization: Artificial Intelligence, Fault-tolerant and Reconfigurable Systems, Data Fusion Thesis: Optimizing Commercial Maritime Port Operations through High Level Information Fusion
• Dynamic algorithm selection and deployment, process optimization, data fusion
• Multi-objective Evolutionary Algorithms, Artificial Life, Neural Networks, Fuzzy Systems, etc.
• Concept Learning Systems and Machine Learning; benchmarking and comparison metrics Optimising Commercial Port Operations Through High-level Information Fusion
International Journal of Logistics Systems and Management Oct 2021
In order to remain profitable, commercial maritime ports must maintain high throughput of inbound vessels. The gantry cranes that load and unload the vessels are the primary point of interaction between a vessel and the port, which cause a critical bottleneck in the process flow. Errors in this segment of the process cause cascading delays which ultimately cause vessel service backlogs, extending to logistical delays in moving shipping containers across land and rail as well. This is to the detriment of the ports, which lose popularity among shipping lines and may even be fined for causing sub-optimal delays. This work expands on prior work in using a multi-objective genetic algorithm to optimise the parameters of a fuzzy system which controls port-side resource deployment. In contrast to existing solutions, this resource deployer is able to function online and adapt to real-world faults while still maintaining superior performance as compared to industry practice. Further, proposals to expand or reduce the port-side infrastructure are computed. Optimizing Maritime Vessel Service Time with Adaptive Quay Crane Deployment Through Level 4 Hard-Soft Information Fusion 22th International Conference on Information Fusion (FUSION), Ottawa, Canada July 2019
Commercial maritime ports must maintain high service throughput in order to remain profitable. One of the most critical operations in a commercial maritime port is the loading and unloading of shipping containers on a vessel (i.e. servicing a vessel) and storing them in the storage yard. A delay in this process would cause cascading delays in servicing further vessels, causing delays in moving cargo across land, rail, and sea. Furthermore, the port itself may incur fines for allowing such delays in their operational procedure. This work highlights a Fuzzy System optimized by a Genetic Algorithm to adaptively control the deployment of quay cranes (and their operators and all other supporting equipment and personnel) to optimize the time required to service a vessel, while simultaneously reducing the operational costs of doing so. Prediction of Container Damage Insurance Claims for Optimized Maritime Port Operations
Canadian Conference on Artificial Intelligence, Toronto, Canada May 2018
A company operating in a commercial maritime port often experiences clients filing insurance claims on damaged shipping containers. In this work, multiple classifiers have been trained on synthesized data, to predict such insurance claims. The results show that Random Forests outperform other classifiers on typical machine learning metrics. Further, insights into the importance of various features in this prediction are discussed, and their deviation from expert opinions. This information facilitates selective information collation to predict container claims, and to rank data sources by relevance. To our knowledge, this is the first publication to investigate the factors associated with container damage and claims, as opposed to ship damage or other related problems.
Master of Science (M. Sc. Computer Science)
University of Ottawa, Ottawa, Ontario
September 2011 – June 2013
Specialization: Artificial Intelligence – evolutionary algorithms Thesis: A Hybrid Genetic Algorithm and Evolutionary Strategy to Automatically Generate Test Data for Dynamic, White-Box Testing
Preliminary results published in IEEE CEC, 2013
• Combinatorial and Evolutionary algorithms (Simulated Annealing, etc., Genetic Algorithms, etc.)
• Formal software design and proof, testing methodology (self-specifying and error checking code)
• Natural Language Processing
INVITED WORKSHOPS
- Effective use of authoring, project management tools to optimize collaborative research and publication
- Time management, burnout detection, mitigation and recovery, to optimize consistent throughput
- Overview of metaheuristic search methods, time complexity, development, testing, and debugging
Contact this candidate