About me

Enthusiastic person with 4 years overall IT experience.
Familiar with modern AI/DL spheres.
Lead the team that won Intel OpenVINO hackathon.
Proactive team member and creative thinker.

My key soft skills are:
  • Problem-solving
  • Adaptability
  • Critical thinking

Experiences

Computer Vison Engineer

2020 - Present
Petcognito
  • Creation and training of DL models
  • Establishment of models training pipeline
  • Model launch on mobile device
  • Development of REST service for model demo

Data Science Engineer

2019 - 2020
Harman, Nizhny Novgorod
  • Creation and training of time series prediction models
  • Unstructured data analysis
  • Research on detection and recognition computer vision models
  • Preparation of demo on embedded devices

QA/DevOps Intern

2017 - 2019
Intel, Nizhny Novgorod

Worked in a cross-geo team.

  • Supported team software development and testing processes
  • Collaborated to product porting on Linux systems
  • Continuous Integration system establishment
  • Take part in team management

Projects

Bachelor's thesis — ASR methods research (using DeepSpeech to recognize speech in Russian language) - Development of the Russian language speech recognition system based on DeepSpeech architecture. The best WER demonstrated by my approach currently equals to 15.8% with language model usage, but could be improved with bigger dataset.
Voice messages recognition bot for vk.com - Bot allows to convert any voice messages to text output. (forwarding the message is the only required action)
Text bot for vk.com - Text bot that sends to users current wearher, traffic jam and much more

Hackathons

I often participate in and win hackathons as a leader of Team Pepe.
My responsibilities include but not limited to AI models creation, product presentation and team management.

  • Intel OpenVINO Hackathon (click for habr post)
  • 1st place
    Speaker&emotion recognition on Raspberry Pi w/ Intel NCS 2
  • TenderHack
  • 3rd place
    Visual search of goods and tenders
  • Sibur Challenge (click for demo)
  • 3rd place
    Chemical production predictive model
  • Ifest (2018)
  • 2nd place
    Room type determination based on a photo
  • Timebook Hackathon
  • 1st place
    Face detection

    Hard Skills & Proficiency

    Python (AI libraries stack)

    Operating Systems

    ML & DL

    Bash

    Algorithms

    Raspberry Pi