Software Engineer Data Scientist

DACOSTA YEBOAH

(***) *** - ****

Houston Tx

***************@*****.***

www.linkedin.com/in/dacostayeboah

Education

Missouri State University, Springfield, MO

Master of Science in Computer Science Aug. 2019 - Jul. 2021 Professional Experience

Senior Data Scientist & Software Engineer, HSR.health Dec. 2021 - Present

● Developed an agent-based model to simulate the spread of diseases in various locations, taking into account factors such as population density, mobility patterns, and environmental conditions. The model has been used to inform public health policies and interventions. Tools used were AWS Glue Crawler, AWS Cloud Formation Template, AWS Event Bridge, AWS Cloud Watch, AWS Lambda, AWS Step Functions, AWS Athena, AWS ECR, GitHub Actions, SIER Model, Agent Based Modelling, Mesa, Geopandas, Shapely.

● Developed a comprehensive drought risk indicator that combines various data sources, including meteorological, hydrological, and socioeconomic factors, to provide a holistic assessment of drought risk in different regions. The indicator has been adopted by several organizations for drought monitoring and early warning systems. Tools used were AWS SageMaker GeoSpatial, R for data processing, Geopandas, ArcPy, ArcGIS Pro, ArcGIS Online, ArcGIS Dashboard.

● Implemented and deployed a Geonode server that allows for the storage, management, and sharing of geospatial data within the organization, streamlining data access and collaboration among team members and stakeholders. Tools used were AWS Cloudshell, AWS EC2, Docker, Docker Compose, Django, PostgreSQL/PostGIS, Celery, RabbitMQ, Jenkins, NGINX, Shell Scripting.

● Developed a cutting-edge system that integrates LLMs with geospatial data and analysis capabilities, enabling users to interact with the system using natural language queries and receive insights and recommendations for public health decision-making. The system has been well-received by public health professionals and has demonstrated the potential of LLMs in the geospatial health domain. Tools used were AWS EC2, AWS Incognito, AWS OpenSearch Serverless, AWS Elastic Beanstalk, AWS DynamoDB, AWS Lambda, HuggingFace Sentence Transformers, Langchain, Spacy, Retrieval Augmented Generation, Django, Streamlit, Tailwind CSS.

Software Engineer, Qremia Evolution Jan. 2016 - Aug. 2018

● Built and maintained RESTful APIs using Django and Flask for the company's SaaS products, enabling seamless integration with third-party systems and enhancing the overall user experience. Tools used were Python, Django, Django REST Framework, Flask, Flask-RESTful, PostgreSQL, AWS (EC2, RDS, S3, CloudFront), Docker, Git.

● Developed a scalable and efficient web scraping pipeline to extract relevant company data and enrich existing lead records with additional details, improving the effectiveness of sales and marketing efforts. Tools used were Python, Beautiful Soup, Scrapy, Selenium, Pandas, Proxies, Rotating IP addresses, MongoDB.

Software Engineer, Process & Plant Automation Sep. 2018 - Jul. 2019

● Developed a voice-controlled smart home automation system that allows users to control various aspects of their homes. Tools used were Node.js, Express.js, TypeScript, Python, Django, Smart Bus Protocol.

● Implemented a fully functional JavaScript framework compatible with the S-BUS protocol, exposed through a well-documented API. Tools used were JavaScript, TypeScript, TCP/IP. Research Scientist, Computational Learning Systems Lab Aug. 2019 - Aug. 2021

● Developed a clustering algorithm to cluster patients with brain injury into meaningful groups, with the potential to help clinicians in making more informed decisions. Published award-winning scientific papers in reputable journals. Tools used were Python, TensorFlow, PyTorch, Keras, AWS Cloud, Django, Unsupervised Learning, OpenCV, Medical Imaging, R, Deep Learning, Scikit-Learn, Matplotlib, Django REST Framework, React.

● Developed a data processing framework compatible with machine learning algorithms, enabling seamless integration of injected data with developed machine learning algorithms. Tools used were Python, Scikit-Learn, AWS Cloud, Pandas, Matplotlib, Seaborn. Published Research Papers

● An Explainable and Statistically Validated Ensemble Clustering Model Applied to the Identification of Traumatic Brain Injury Subgroups (Published in IEEE Access, 2020)

● Statistical Comparative Analysis and Evaluation of Validation Metrics for Clustering Optimization (Published in IEEE SSCI, 2020)

● Heterogeneity in blood biomarker trajectories after mild TBI revealed by unsupervised learning,

(Published IEEE TCCB, 2021)

● A PheWAS Model of Autism Spectrum Disorder (In press, IEEE TCBB, 2021)

● A Deep Learning Model to Predict Traumatic Brain Injury Severity and Outcome from MR Images (IEEE CIBCB, 2021)

Awards

● Best paper award, IEEE CIBCB2021 - A Deep Learning Model to Predict Traumatic Brain Injury Severity and Outcome from MR Images.

Contact this candidate