Staff Software Engineer2019 - PRESENTQualcomm
Technologies: C++, Algorithms, Data Analysis, Cloud, Pandas, NumPy, Perforce, Scikit-learn, Ggplot2, Ubuntu Linux, C, Plotly, Tidyverse, Keras, Eclipse, Bash, Shell, Data Mining, Data Modeling, Data Analytics, Data Visualization
- Developed an automation pipeline using Jenkins and Python to run continuous simulations, process and clean results, store them in SharePoint and MySQL using Python API, and visualize results using Plotly.
- Created and adapted complex machine learning algorithms, models, and frameworks aligned with product proposals or roadmaps.
- Tracked and fixed bugs using Jira as a reporting tool. Improved debugging and research skills by finding the root cause of complex issues.
- Enabled and optimized state-of-the-art neural network models to meet the demands of customers' real-world use cases.
- Developed innovative data analysis and visualization tools.
Machine Learning Consultant2015 - PRESENTSelf-employed
Technologies: Artificial Intelligence (AI), Machine Learning, PyTorch, TensorFlow, Natural Language Processing (NLP), SQL, Recommendation Systems, Data Science, SciPy, Deep Reinforcement Learning, Neural Networks, RStudio Shiny, Computer Vision, Data Visualization
- Deployed machine learning code, models, and pipelines into production and troubleshot issues that arose.
- Built a first-class machine learning platform from the ground up, which helps manage the entire model lifecycle, including feature engineering, model training, evaluation, versioning, deployment, online serving, and monitoring prediction quality.
- Employed machine learning and statistical modeling techniques, such as decision trees, logistic regression, Bayesian analysis, and neural networks to develop and evaluate algorithms to improve product and system performance, quality, and accuracy.
Senior Software Engineer2015 - 2019Blue Danube Systems
Technologies: Python, MATLAB, Simulations, Random Forests, Reinforcement Learning, Deep Learning, R, Docker, PySpark, Tableau
- Created a Flask-based web application to simulate the signal received by the user in a cellular system and visualize the data on Google map.
- Designed and developed software for simulating complex wireless networks in Cpp and MATLAB.
- Developed deep reinforcement learning models and deep neural networks, including Graph NN, CNN, RNN, and attention and transformer.
- Designed and developed an automation pipeline to extract user and network KPIs, store data in a MySQL server, preprocess in Python, and visualize the results in Tableau.
Research Assistant2011 - 2014Virginia Tech
Technologies: Artificial Intelligence (AI), Machine Learning, Optimization, Scraping, Web Scraping
- Developed a feedforward neural network model to predict users' location using time-of-arrival data.
- Compared the proposed model with the state-of-art solutions regarding running time and performance in terms of root mean square error (RMSE).
- Developed a web application using Flask that compares the prediction of different machine learning models for Node.js localization based on user input data.