Rinat Mullahmetov

Rinat Mullahmetov Email and Phone Number

Data Science @ Звук
moskva, moscow, russia
Rinat Mullahmetov's Location
Innopolis, Tatarstan, Russia, Russian Federation
About Rinat Mullahmetov

Master Degree at Innopolis University Data Science Program. Experience in solving problems at all stages of a Data Science project's life: from data analysis and markup to model training and deployment of the final solution.

Rinat Mullahmetov's Current Company Details
Звук

Звук

View
Data Science
moskva, moscow, russia
Website:
zvooq.com
Employees:
35
Rinat Mullahmetov Work Experience Details
  • Звук
    Data Scientist
    Звук Nov 2023 - Present
    Innopolis, Tatarstan, Russia
    * Design and develop an LLM based solution for generating playlists based on a text query (HF, Transformers, Embedding Learning, LORA, Accelerate, MTEB)* Evaluate this solution* Deploy using Triton Inference Server
  • Innotech
    Data Scientist
    Innotech Jun 2021 - Nov 2023
    Иннополис
    1) DatasetGAN research in the task of generating a dataset of vehicle parts segmentation2) Vehicle Damage Detection system:- Details Segmentation- Pose Estimation- Angle Prediction- Damage Instance Segmentation3) Designed and implemented a benchmark for evaluating the outputs of models and different versions of training sets4) Development of an antifraud system for assessing damage from street cameras- Design and train models using OpenMMLab and Detectron2- Export models to ONNX/TRT- Design and implement production pipeline using Triton Inference Server:* Preprocessing using DALI backend* Host models using ONNX/TRT/Python backends* Implement business logic using Triton BLS logic* Construct pipeline using Trition Ensemble
  • Университет Иннополис
    Data Scientist
    Университет Иннополис Apr 2020 - Jun 2021
    Россия
    1) Developed Chest-X Ray analysis system based on Nvidia Clara Deploy2) Developed Chest X-Ray View Position Classification Model3) Designed and implemented a Chest-X Ray analysis system, which outperformed Clara based solution.The system is recognized as the leader of experiment 2020
  • Eora
    Data Scientist
    Eora Oct 2018 - Apr 2020
    1) Investigated classical methods of identifying descriptors to solve the problem of finding similar logos. Developed a SIFT-based solution and integrated it into the existing search algorithm of a trademark search service2) Participated in all stages of the development of a service for recognizing water meters. (selection and markup of data, development and training of models for classification of numbers and flip, combining models into a pipeline, writing a back end, containerizing an API, developing a client in the form of bot telegrams.)3) Participated in the development and implementation of a model for identifying features from an image for a similar logo search service4) Participated in the development of a pipeline for video styling. Solved the problem of interior segmentation, implemented a “ribbed” colourization of furniture and human clothing. Implemented the 2d facial landmarks model and added the "glasses" effect based on them

Rinat Mullahmetov Education Details

Frequently Asked Questions about Rinat Mullahmetov

What company does Rinat Mullahmetov work for?

Rinat Mullahmetov works for Звук

What is Rinat Mullahmetov's role at the current company?

Rinat Mullahmetov's current role is Data Science.

What schools did Rinat Mullahmetov attend?

Rinat Mullahmetov attended Университет Иннополис, Университет Иннополис.

Who are Rinat Mullahmetov's colleagues?

Rinat Mullahmetov's colleagues are Влад Митюкляев, Ульяна Нижникова, Ilya Dalechin, Ermek Aitygulov, Evgeny Ronchinsky, Vladislav Averin, Alexey Kott.

Free Chrome Extension

Find emails, phones & company data instantly

Find verified emails from LinkedIn profiles
Get direct phone numbers & mobile contacts
Access company data & employee information
Works directly on LinkedIn - no copy/paste needed
Get Chrome Extension - Free

Aero Online

Your AI prospecting assistant

Download 750 million emails and 100 million phone numbers

Access emails and phone numbers of over 750 million business users. Instantly download verified profiles using 20+ filters, including location, job title, company, function, and industry.