- Led a team, with the CEO and key stakeholders, to assess and extract knowledge graphs from published documents for corpus-constrained search and retrieval.
- Independently identified inefficiencies in analyzing digital risk. Developed a python-based language-agnostic topic analyzer, reducing analytical delivery times by 80%.
- Improved output fidelity with RLHF inferencing to incorporate domain expertise.
- Championed and enforced best practices across the software development lifecycle, implementing CI/CD with
- GitHub Actions and docker containers.
- Collaborated directly with customers and stakeholders to architect and launch a city-scale agent-driven simulation, visualizing vaccination rollout.
- AWS
- Terraform
ansible
nginx- GitHub
- GitHub Actions
- Python
- LLMs
Streamlit
Airflow
- Designed, developed, and evaluated ML classifiers to improve streaming device conversions.
- Segmented customers by purchase propensity using audience segmentation techniques.
- Imputed incomplete user data with sampling and interpolation, improving efficacy.
- AWSSagemaker
- Hive
- GitHub
- Created and taught curriculum for using Python in ETL workflow integration
- Conducted lab demonstrations for students.
- Filled in when the professor was unwell.
- Held office hours to answer questions.
- Graded student assignments and labs.
- Courses:
- Python
- Java
- Engineering Ethics
- Algorithm Design
- Blackboard
- BrightSpace
- Evaluation
- Teaching
- Created and delivered lectures and demonstrations to teach Python and algorithms.
- Developed rigorous evaluation methods, and managed teaching assistants.
- Implemented plagiarism detection and handled plagiarism cases.
- Evaluation
- BrightSpace
- Developed and ran multi-objective optimization simulations for resource scheduling using MATLAB and C
- Extensive use of optimization algorithms (Non-dominated Sorting Genetic Algorithm, etc)
MATLAB- C
- Evolutionary algorithms to predict imapact of time-series events (improved prediction accuracy by 36x)
- Co-inventor on a a patent application relating to modeling attribution of advertisement features
- Genetic Algorithms
- Behavioral Analytics
- Developed automated test scripts in Python to test wi-fi capabilities of BlackBerry handhelds and tablets
- Developed a modular framework to simplify future development of such test scripts
- Improved efficiency of test case analysis by over 200x
- OmniPeek
- OmniEngine
- Agilent Wireless Technologies
Wireshark
- Developed web applications to display on mobile and desktop browsers.
- These applications served general administrative and sensitive student information.
- Tested user interfaces for usability and cross-platform compatibility.
- PHP
- MySQL
- Developed automated testing framework for the company’s premier product line.
- Designed and imlemented tests for pre- and post-conditions; along with rollbacks and rich logging.
- Integrated Python with Microsoft’s .NET framework.
- Set up virtual development environment for future interns, to improve commissioning latency.
- Python
- Microsoft .NET
- Analyzed web traffic and AdWords campaign and explored other online advertising avenues.
- Designed a program to rank URLs to which to add back/links to improve Google Page Rank.
- Python
- SEO
- Page Rank
Dissertation: Optimizing Commercial Maritime Port Operations through High Level Information Fusion
- Developed algorithms and implemented advanced data fusion techniques for dynamic selection and deployment of algorithms and datasets to optimize maritime operational processes.
- Applied Multi-objective Evolutionary Algorithms, Neural Networks, and Fuzzy Systems.
- Designed evaluation metrics for Machine Learning models.
Dissertation: A Hybrid Genetic Algorithm and Evolutionary Strategy to Automatically Generate Test Data for Dynamic, White-Box Testing
- 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
Awards - 1st place, uOttawa Research Poster Competition (photo) - 2nd place, IEEE Poster Competition (photo)
Links - Poster
- Artificial Intelligence – learning, planning, exploration, adversarial game logic, neural networks Strong foundation in predicate logic, semantics analysis and related theory
- Data structures, algorithm design and optimization
Evolutionary algorithms are a class of algorithms that try to mimic natural, biological evolution a la Darwinian natural selection, to compute solutions to a given problem. They are especially useful when no well known strategies for computing solutions to such a problem exist. Evolutionary algorithms begin by creating a collection (population) of candidate solutions to the problem at hand; and through repeated application of genetic operators such as crossover and mutation, they iterate over multiple generations of this population, until they eventually converge onto an attractive solution. One important problem facing code implementing Evolutionary Algorithms is that due to the dynamic nature of the individual chromosomes in a population, simple coding errors lead to complex bugs that are difficult to both diagnose and debug. This problem is only exacerbated when attempting to develop the algorithms in a dynamically typed language such as Python. This paper presents a novel Evolutionary Algorithm framework for the Python programming language that implements design-by-contract, a paradigm in which each function and class must follow a contractual set of pre-conditions and post-conditions. Failure to follow the contract causes an error condition identifying the violated clause, thereby catching bugs earlier in the development process and in a more descriptive manner.
Links - Paper - Official Documentation - Discussion and Motivation - Presentation slides - On PyPI
A pure-Python decorator-based design-by-contract framework.
A lightweight developer tool to add inline comments into application logs.
Real-time capacity management in healthcare scheduling in a multi-tenant architecture.
Using Python's NLTK module and WordNet/PyWordNet to accurately describe an event recorded by multiple text documents.
Links - Proposal - Presentation slides
Artificial Immune Systems (AISs) are intelligent methodologies inspired by the immune system towards real world problem solving. They are systems that lend themselves easily to meta-heuristic algorithms to solve problems more efficiently in high volume and high dimensional solution spaces. Inspired by the mechanisms that control the human immune system, AISs perform especially well in pattern recognition tasks — an ability that is used in solving multiple different types of problems. With the use of techniques from other domains of meta-heuristic programming, such as Genetic Algorithms and probabilistic/heuristic searching, the range of problems to which AISs can be effectively applied is greatly increased.
Links - Presentation slides