Anthony Polyak Email and Phone Number

‣ Designed Django backend for full stack “BarHub” application, feature rich and functional - developed “battle” mechanics to adjust and track ratings complete with leaderboards and win/loss records, moderators able to both judge / report battle outcomes via Django ‣ Employed Pymongo/Djongo package to utilize MongoDB Atlas for highly available, low latency and high throughput database access‣ RESTful API provides flexible access to user data and likewise provides significantly greater control in display and implementation of data, enables integration into external applications including BarBot allowing display of statistics and ratings for each user on demand‣ Full featured user account customization, Oauth2 integration with Discord, persistent personal ratings and performance metrics

‣ Developed a “CypherBot” to facilitate Freestyle Cyphers including moderation and live generation of games and words, Integrated Datamuse API for random word generation, intuitive user selected parameters define live-generated list using word quality/popularity ‣ Implemented MongoDB as the backend database for user and server settings to enable remote server deployment and persistent server-wide feature toggling, storage of user settings and parameters and persistent storage of user data, leveraging MongoEngine ODM to provide non-blocking code execution for maximally responsive command execution and constantly available delivery of data‣ Interactive user elements for providing feedback / suggest feature improvements and wordlist optimizations, surveyed user feature requests and implemented requested features to randomize and display games and challenges with various user specific parameters‣ Featureful user settings system allowing for the enabling and disabling of pings, reactions, message embedding, customizable intervals for generation, number of words generated, type of words generated, delay time, gamification parameters and scoring‣ Database interface structure using Object Relational Mapping (ORM) Class based framework for DRY design, fast and highly available database access based on object attributes, each object containing all required data for storage and retrieval of necessary information‣ Implemented asyncio for Asynchronous communication with Discord.py for command design pattern and integrated Oauth2 Discord authentication system for permissions and user levels as well as object relational integration into database information storage‣ Rebuilt BarBot and packaged requirements for deployment to local Raspberry Pi server using object oriented compartmentalization of features using cogs - modularized class based fault tolerant and resistant structure enabling high uptime / stable delivery of services

Investigated system failures under high workloads. I determined the problem was related to exceeding managed-service capacity. I then designed and developed a solution utilizing shared connections. Results were 90% drop in resource utilization and cost, and eliminating system errors. Solution was developed with Node.js and was adopted and successfully utilized by all services accessing Azure CosmosDB.Successfully performed complex and urgent incident response activity. Pioneered first definition of incident response operations, including documentation and training of other engineers. On-call activity required ensuring compliance with SLA and coordination with experts across the enterprise.Member of a specially selected 5-person team working to load and transform many data formats from many sources into a common data format for business analysis. In addition, load time was exceeding available process time. We created a solution which reduced load time from 2 weeks to 2 hours (per file) with efficient algorithms and distributed parallel processing. Personally responsible for designing and building the most efficient and high performance solution for the scraping and splitting components of the solution. Technology used was Python and Selenium.Identified four areas needing improvement across development teams. Consistently advocated and provided improved tools and processes for the four areas. The four areas were documentation, coding standards & linting, integration testing, and available test data. Worked directly with Microsoft Azure Engineers to diagnose and optimize solutions for issues ranging from database connectivity and resource utilization to firewall and service configurations.Personally responsible for developing external calling software. Also designed and developed internal rate service systems.Technologies utilized were: Java, Spring, Maven, JavaScript, Node.js, Express.js, GraphQL, REST.

‣ Leveraged FuzzyTools to process and remove similar words within a user defined or automatically incrementing tolerance ranges depending on word length and word complexity ratings and returning customized word-matched lists for user verification of accuracy‣ Used Itertools to create efficient combinatorial pairings for fast / efficient word processing using FuzzyTools similarity rating indexes‣ Automatically generated user customized ‘cleaned’ lists for the purpose of display and integration with BarBot word generation