I have over 4 years of experience developing scalable, cost-efficient, fault-tolerant solutions using Amazon Web Services. I am AWS-certified on building event-driven serverless architectures on their platform.
Services: Pipelining, data storage/migration/optimization, ETL logic, orchestration, infrastructure-as-code, model training and serving, architecture optimization, CI/CD implementation.
Past Work: Architecture was event-driven, scalable, and fault-tolerant. It was developed using infrastructure-as-code principles. CI/CD of the code repository was managed through GitHub Actions. Solution utilized the following AWS services: IAM, EventBridge, S3, RDS, ECR, EC2, Lambda, Cognito, CloudWatch, CloudFormation
If you want to engage in a completely new data initiative, I can evaluate feasibility and build anything from basic proofs-of-concept to fully-functional data science MVPs. If you'd like, I can plug those in to your current technology stack. Or if you have a technology you prefer working with, I can build something using it, or something that is compatible with it.
If you desire to move from concept towards full deployment, I can host these solutions using cloud services that make it affordable to serve thousands of users. When my clients have decided to migrate to production-grade solutions quickly, my codebase is built using infrastructure-as-code practices and can easily be migrated or harvested into your new stack.
Services: Architecture design, feasibility scoping, market research, proof of concept software development, minimum viable product software development, core feature development.
Past Work: I have built sample dashboard and pipeline setups for clients that ingested a subset of their desired data as an Excel file, processed that data, and produced some simple interactive visualizations. From there, I migrated this dashboard to AWS hosting, added login credentials and secure IP subnet, and built a custom URL handler they could use to interact securely with stakeholders for feedback.
I have been building pipelines for years and can make them do whatever your business needs, cost efficient and low latency. Any unique requirements can be handled such as significant batch processing scheduled daily, and pipelines that require low latency, real-time predictive analytics.
Services: Batch processing, real-time processing, model registry and serving API endpoints, auto-scaling, ETL logic.
Past Work: I built a Data Lake and pipelining infrastructure that was capable of ingesting virtually any form of written text, processing the text, conducting feature engineering, and loading the resulting insights into the application backend for visualization. The pipeline was built in AWS and performance was tunable, scalable, and fault-tolerant.