· Created an ETL pipeline to ingest data from several sources to store in a SQL database · Integrated a Sentence BERT model trained on patents to encode patent text data for similarity searching · Trained a FAISS index for quick comparison of over 2.3 million claims text, ranked by cosine similarity · Designed and wrote a Streamlit frontend application for easy accessibility and saving hours to under 30 seconds

· Wrote dbt SQL jobs to transform data on 100+ TB Snowflake data warehouse· Created Prefect tasks and workflows to process and interact with Prefect API for flow run metadata daily· Identified long running and expensive dbt jobs, Snowflake queries and Prefect flows for improvement· Optimized SQL queries, reducing runtime from over 30 minutes to under 2 and saving over $11000 per year · Handled interaction between Snowflake, AWS S3, Perfect and dbt jobs

- Searched, evaluated and imported data in different formats from various sources worldwide - Studied and implemented ways to improve database, such as using mathematical concepts for company disambiguation - Integrated client data - Analyzed user and performance metrics for areas of improvement- Communicated and worked with Sales team to ensure potential client's questions are answered and their needs are addressed- Integrated third party APIs for mass data imports- Managed automated crawling of over 1200 websites- Administered database with over 100 million publications, grants and patent records and millions of people and company profiles

- Designed and wrote a standalone proprietary program with user interface to automatically calculate potential new product sales based off competing benchmarks - Created and compiled a database of 16,000 competing products spanning across over 100 brands through web scraping and merging data provided by parent company - Participated in the development life cycle of new products to understand the concept and key aspects of product - Examined market competitor for new products and their performance- Analyzed and visualized industry data based off specific product categories and criteria - Forecasted potential sales of future product launches based on internal and industry data

-Assisted in teaching programming with Python to students who have no programming experience.-Held office hours to help students with homework and exam preparation in private.-Learned how to give constructive feedback through grading of students’ works.

-Acquired experience working with big data, data mining and programming software (Visual FoxPro).-Identified mistakes and incomplete data within a database of above 170,000 transactions and cleaned it through SQL.-Created a program that reads a number into an alphabetical sentence up to hundred billion.-Developed business relationship, both within the office and interaction with clients and business partners.-Trained customer’s employees to use custom-made software.

-Conducted a feasibility study on the practicality and possibility of using 3D printing to manufacture small car parts.-Gave a presentation to managers on the aforementioned study and along with ways to utilize the 3D printer in more ways than printing car parts such as a customized memento for each customer.-Worked on customer service and accepting work orders and reservations.-Conducted surveys across hundreds of existing customers as well as compiling results of previous surveys spanning across 4 months.