Working with market.Developing strategies, backtesting hypothesis, improving trading tools (+ developing backend and front-end before backend was hired), using ML for time series prediction/analysis/generation.

Summary: successfully built (key dev) and launched NN lung chest x-ray pathology classification model in Moscow clinics (the best in 2020), team-lead of 5 people in ML projects, teamlead and pm of labelling tool project, had plenty of successful pre-sales in medical and industrial tasks, helped with other projects. Promoted model version control in lab. Managed chaotic projects and set them on rails.More detailed projects:Definition: ML Team-Lead (Medicine)Performed: Managing team of Junior + Middle ML Engineers, managing hiring process + interviewing, setting hypothesis, pipeline verification, doctor/dev connection, obtaining data, labeling management, data analysis and simplification for further development, mentoring, resource management, promoting MLOps.Definition: Lung pathology classification (chest x-ray).Performed: Built a training classification pipeline that makes use of different kinds of data: images, pathological labels and demographic information. This model was top-1 in 2020, according to mosmed.ai. Led further development and integration it in production pipeline. Made further improvements according to doctors' requests.Definition: Hematoxylin and eosin slice classification/segmentation (histopathology).Performed: Wrote a project within a week for pre-sale. Signed a collaboration with clinics. Worked in collaboration with them.Definition: Aeroflot - engine blades defect detection.Performed: Applied Detectron2 with modified preprocessing to detect defects of aircraft engine blades. Additionally, I wrote a special inference that was later used on this project by others.Definition: Team lead of labelling tool developmentPerformed: managing team (and product development itself) of front-end and back-end developers in development of labelling tool for medicine. Additionally, managed refactoring and fixing of existing demo tools. Extra: wrote few functions for front-end in order to ease further development (related to cornerstone.js).

Working with tumor classification and segmentation-Developed framework that is used for multifile classification with a lot of parameters (data balancing ratio, data balancer, feature seletor, heatmap parameters, classification methods selection)**Framework is tested and was used to write article (accepted) for CIBB conference in 2019. Then used to write extended version of the paper.**It was also used to write another paper for breast cancer (for medical journal, waiting for journal response)-Working on computer vision tasks--3D volume segmentation in medicine---Developing new models based on existing ones as well as creating new models, because 3D volume segmentation is not that popular as 2D---Segmentation of volumes based on 2D models with further processing by 3D model for better segmentation results-Image classification (prostate, breast, biopsy) to reduce burden of doctors--Applying straight transfer learning to existing data--Developing new models based on TL. E.g. developed new fused model where we have two parallel path that merge later. This increased convergence speed as well as final accuracy scores.

Assisting hardware and software problems,Configuring new PCs,Solving network problems.Working remotely and locally.

• Wordpress---Crypto converter with memory (front- and back-end)---Interactive multilanguage site for diving company• Landing pages with adaptive layout---Single page store---Wall map generator---Gallery page with lazy loading, settings, browser-tree view• React.JS apps---React + Redux blog (more details in projects section)---My web page (WIP) is built with react. It’s main structure and logic was used by my colleague to write her project.---Assisting in writing of GeoCalculator that uses both front-end and back-end to solve complex mathematical problems---Weather app• Other apps---Websocket snake games---One-page app using REST APIs

Assisting professors with practical classes,Helping students developing their applications in classes.