Dimcho Karakashev

· Toronto, ON, Canada · dkarakashev at protonmail.com

I am a Software Engineer with full-stack development experience and expertise in the e-commerce domain. At SSENSE, I was part of the Shared Services and Developer Experience team, where I developed highly scalable microservices and serverless applications, as well as internal tooling, and standardized platform processes. Most recently, I joined the Data team, where I drove efficiencies and developed Generative AI solutions.

Previously, I graduated from the University of Waterloo, where I completed my Master's degree under the supervision of Professor Srinivasan Keshav and Professor Sergey Gorbunov. In my thesis, I combined security, privacy, and cryptographytopics that I am very interested inwith climate changean issue that I am passionate about. Before my Master's degree and during my undergraduate studies at Purdue University, I researched under the supervision of Professor Hong Tan. I focused on Haptics/Virtual Reality and created two prototypes involving both hardware and software. The goal was to understand how Haptics can improve the immersive experience in Virtual Reality.


Experience

Software Developer 3

SSENSE · February 2022 - Present
Data
  • Developed a Generative AI agent to automate issue investigations using AWS Bedrock and AWS Strands for the team, reducing investigation time from over 10 minutes to under 1 minute for common alert patterns
  • Led the redesign and implementation of machine learning model blue/green deployment using SageMaker and reducing infrastructure footprint leading to savings of over 30% per month in SageMaker real-time endpoint spending
  • Identified a critical scaling bug prior to production release by evaluating the performance of a new recommendation system, ensuring the success of a project critical to company initiatives
Shared Services and Developer Experience
  • Implemented GDPR (General Data Protection Regulation) user data deletion automation by developing a custom internal proxy and integrating it with OneTrust, boosting efficiency and saving one developer-day per week.
  • Partnered with senior developers and architects to develop both the front-end and back-end of an internal developer portal using React, Typescript, and Backstage for efficient cataloging of over 224 projects.
  • Engineered a pubsub service using AWS CDK and Python, capable of processing over 1000 requests per second while maintaining 99.999% uptime SLA.
  • Designed and implemented company-wide security processes for comprehensive Kubernetes and serverless secret management using AWS Secret Manager, enhancing security by managing over 95% of all company secrets.
  • Enhanced key user features of the internal developer portal by integrating Redis and redesigned asynchronous behavior, reducing latency by over 98% from 5 seconds to 100 ms and improving user experience.

Software Developer 2

SSENSE · January 2021 - February 2022
  • Co-developed a tool for on-demand scaling of microservices, achieving over 10% reduction in Kubernetes costs.
  • Developed multiple Python-based automated tools addressing company-wide technical debt, such as automated cleaning of 146 Feature Flags.
  • Implemented automation tool for Datadog monitor creation and Datadog dashboard enhancement increasing the operational visibility and cost efficiency of services.
  • Worked across 4 teams in cross-functional business areas to resolve 12+ JavaScript performance testing issues preventing impact to the core service's SLA over bot attacks

Software Developer

Contractor for a company based in the USA · Summer 2020
  • Collaborated on designing an object-oriented application for user authentication via smart card.
  • Developed the security-related software of the application using C# capable of multiple factors of authentication to provide the option of balancing security and usability.
  • Documented my contribution to facilitate the maintainability of the application.

Haptics Intern Engineer

TangerineX · Summer 2017
  • Gave weekly updates to the supervisor of the project and self-determined personal tasks and priorities to finish the project on schedule.
  • Updated the firmware of a PCB to add Bluetooth capabilities for wireless interaction with the Oculus application and signals for a realistic tranquil Haptic experience.
  • Presented a demo and plans of the project to a collaborating company and an investor to demonstrate accomplishments which led to additional meetings to discuss possible funding.
  • Managed a colleague by delegating tasks and conducting weekly meetings.

Teaching Assistant(Programming Applications for Engineers)

Purdue University · Spring 2016 - May 2018
  • Lead a two-hour weekly lab of approximately 27 students.
  • Evaluated labs and homework assignments in a timely manner.
  • Provided adequate feedback to the students and explained concepts to improve students’ understanding.
  • Collaborated with about 15 UTAs to solve any issues such as scheduling conflicts.

Education

University of Waterloo

Master of Mathematics
Computer Science - Focus: Security/Privacy

GPA: 88/100

September 2018 - September 2020

Purdue University - Main Campus

Bachelor of Science
Electrical and Computer Engineering

GPA: 3.94/4.00

August 2014 - May 2018

Projects

Thesis: Secure Trading of Renewable Energy Certificates

  • Designed a system to combine a secure resource-constrained device, Azure Sphere, with both permissioned blockchain, Hyperledger Fabric, and permissionless one, Algorand.
  • Implemented high-level APIs using Node.js and low-level applications using C to evaluate the proposed system.
  • Conducted a comprehensive literature review on designing secure smart meters to serve as trusted oracles in the system.
January 2019 - September 2020

Augmented Unlocking Technique for Smartphones

  • Collaborated to develop a new technique for smartphone authentication based on pre-touch information simulated using UNITY engine.
  • Collaborated to design an experiment comparing our technique to previous existing solutions.
  • Discussed the analyzed data and based on preliminary results our technique is more secure with the cost of longer authentication times.
Fall 2018

Hyperledger Fabric Benchmarking

  • Investigated the throughput and latency of Hyperledger Fabric with transaction intervals sampled from Piosson distribution to determine the validity of previous studies.
  • Developed a Python script to automatize the data collection and plotting.
  • Gained a basic understanding of Docker and Docker swarm to explore the impact of different Hyperledger Fabric configurations.
Fall 2018

Stenography - GUI Python

  • Developed a GUI application using PySide in which a user chooses one big image and one small image and embeds the small image into the lower bits of the bigger one.
  • Optimized performance of the initial implementation by a factor of 5 through making full use of NumPy. The project was in the top 10 fastest implementations out of 160 students.
  • Implemented additional features such as dragging and dropping of images and different types of compression.
Spring 2017

Smart Home System

  • Designed a system, including simple circuits, to provide functionalities such as controlling a lamp through a Bluetooth module, interfacing with a temperature sensor, and others.
  • Implemented all the C code for a 9S12 microcontroller to interface with the sensors and the Bluetooth module.
Fall 2016

Club: Remotely Controlled Vehicles

  • Gained knowledge about the process of developing a PCB (Buck Regulator).
  • Learned how to solder PCB parts on a board and how to use a popular software for designing PCB called Eagle.
  • Utilized version control in GitHub to collaborate with all the members of the Electrical Team in order to create parts for libraries, work on schematics, and board layouts.
Fall 2015

Club: The Association for Computing Machinery (SigBots)

  • Collaborated with members in the club to build robots for competitions and to have fun.
  • Modified a proportional-integral-derivative controller of the robot in order to improve the movement of the robot.
  • Learned C and interacted with different sensors such as potentiometer.
Fall 2015

Skills

Programming Languages & Tools
  • Languages: Python, Typescript, JavaScript, Bash, Groovy
  • Frameworks: Node.js, Express, Backstage (Spotify), Jest, Mocha, React, Material-UI
  • Cloud: AWS - SAM, SLS, SNS/SQS, Lambda, Eventbridge, Firehose, CloudWatch, Secret Manager
  • Other: Jenkins, Redis, Git, Datadog, Docker, Kubernetes, VS Code, JIRA, CMake, LaunchDarkly, Prometheus

Awards & Certifications

  • David R. Cheriton Graduate Scholarship, University of Waterloo August 2018 - August 2020
  • International Masters Student Award, University of Waterloo August 2018 - August 2020
  • Distinguished graduate, Purdue University May 2018
  • Summer Undergraduate Research Fellowship, Purdue University Summer 2016
  • Dean's List and Semester Honors, Purdue University Fall 2014 - Spring 2017
  • 1 st place in NASA Space Settlement Contest, San Diego May 2013
  • Found two asteroids in Killer Asteroid Project October 2013