Sebastian R. Estenssoro

Successfully transitioned third-party Docker files to secure internal images, ensuring uniformity across all Kubernetes services. Contributed to large Terraform codebase for managing Cloud Kubernetes Clusters. Authored custom Prometheus endpoint for Cluster Autoscaling metrics in Golang. Created Golang Pulumi Kubernetes CRDs for managing infrastructure

Implemented custom Terraform Providers to facilitate Hybrid Cloud Computing infrastructure, alongside designing and building Terraform modules for a private AWS EKS deployment within the Twitter VPC. Contributed to the successful evaluation of Golang as the primary non-JVM language at Twitter. Developed automation tools in Golang for GitOps and compute infrastructure which played a critical role in enhancing operational efficiency. Reduced the time required to provision the Terraform Enterprise server infrastructure from 4 hours to just 5 minutes. Lead cost-saving initiatives for compute and storage in GCP. Developed a Golang Auth add-on tool for Kubernetes config maps.

IT consulting providing freelance expertise on building and deploying web services

Built and lead an engineering team to deliver enterprise level technical solutions to a multi-billion dollar company. Championed and implemented Cloud Web Services, Agile Development Frameworks, Data Warehousing, BI Reporting Platforms, Containerization, and Serverless Technology.

Designed and maintained numerous data pipelines in Python and Golang. Rewrote and customized the popular workflow program Airflow to run on Windows. These pipelines connected to a variety of databases (Microsoft SQL, OpenTSDB, ElasticSearch, and Oracle) and moved data into a Data Mart and Data Warehouse for analytics and reporting. Built Django Web applications to help users to interface with these databases and data warehouse.

Contract Data Scientist for large telecom company. Worked in a Hadoop/Mapr environment programming data pipelines and algorithms for classification using TensorFlow and SciKitLearn. Set up Hadoop cluster from scratch, and moved large amounts of data between Amazon EC2 instances. Displayed data using Python's Django web interface.

Special data consultant on the Aliso Canyon Gas Leak Disaster. Self-taught Python to streamline the ETL efforts of relocation data. Spearheaded data and analytics and Interacted directly with VPs to provide updates on the relocation efforts and wrote algorithms to verify more that $20M in claims reimbursements.

Created and implemented VBA macro-enhanced verification sheets that were adopted in all Halliburton districts in the Rockies while managing and mentoring a team of 5 engineers. This position required 24-hour availability to support cementing operations in the lab and field. Invented a variable density/viscosity fluid system and programmed data-smoothing algorithms that lead to increased productivity.

Worked directly under the CEO to interpret historical sales data and predict budget and shipment needs. Created and maintained up-to-date market trackers and business reports and self-taught VBA to automate this process. Efforts lead to the replacement of Salesforce with in-house reporting tools.

Web application built using AWS, Golang, GraphQL and Vue which allows users to create custom course profile stickers for endurance sport races

Web application built using AWS, Golang, GraphQL and Vue. This is an annual scavenger hunt at Buttermilk Ski Resort where contestants use QR codes to find scavenger items across the ski mountain and compete for most points

Web application that allows users to create APIs and schemas and drop excel or csv files to be automatically ingested and available via GraphQL and RestAPI. Future plans are to allow connections to other data stores and relationships between schemas

Golang ORM for modeling your data and interacting with SQL. I built this because available ORMs lack the ability write multiple models efficiently. We have largely replaced this with Gorm in our projects but use functions and concepts from Dasorm to improve performance for mass CRUD operations.

There's no denying the ubiquity of the Pandas DataFrame in Data Science. It's simple interface and powerful transformations make it a go-to for most machine-learning professionals. The problem comes when a Data Scientist needs to upload their dataframe to a SQL-like database. Traditionally, a Data Base Administrator will have the Data Scientist drop a csv-file in a predetermined location and set up a trigger and process that will absorb the file into a table.

SQLWriter enables a python programmer to seamlessly upload a Pandas DataFrame to several flavors of SQL database with 3 lines of code.