• Created OkGPT t.me/ok_gpt_bot, an AI assistant bot platform. It allows users to interact with the most advanced AI through voice and text while offering additional productivity features and integrations.• Optimized the code to parallelize user queries' processing to increase the bot performance by magnitude.• Implemented a complex subscription system with multiple tiers, referral links, discounts, and payment options.• Implemented support for most of the languages in text and voice messages.

Toptal Client Projects (highlights):Real Estate: Residential Property Pricing• Developed an end-to-end ML solution to predict real estate prices, eliminating manual assessments and enabling quick customer responses.• Designed an auto-refreshing data pipeline for daily data processing from AWS S3.• Built an interpretable ML model using FBProphet, FAISS, SHAP, and more, meeting business constraints.• Created a historical data simulation system for synthetic data generation and backtesting.• Led data science team members and collaborated with engineering.HR: Talent Sourcing• Led the development of a core ML engine for talent scoring using Databricks and Spark.• Built an ETL pipeline to load a terabyte of raw data into Elasticsearch and Kibana on AWS.• Enhanced the Similarity engine with semantic text embeddings using NLP neural networks.Blockchain Cybersecurity: Rapid Threat Detection• Developed an ML classifier to detect malicious smart contracts automatically.GenAI Detection• Trained a custom BERT neural network to identify AI-generated software code.• Created a suite for generating synthetic labels and data using LLMs.

• Developed and led the creation of innovative courses in Mathematics, Data Analysis, and Machine Learning• Successfully recruited and onboarded skilled lecturers and tutors to enhance educational delivery

• Data Science project lead: data exploration, data pipelines, data quality, data visualization, machine learning models• Bayesian inference, probabilistic graphical models, interpretable models, uncertainty estimates• Client communication• Interfacing KPMG Data & Analytics group with academic science

• Conducted research in Neuroscience, Signal Processing, EEG and MEG analysis• Invented and developed the first algorithm to detect cross-frequency phase coupling between brain rhythms, based on M/EEG• Supervised students• Taught courses: Computational Neuroscience, Programming in MATLAB

Conducted research in: • Dynamical systems, attractors, partial and non-uniform hyperbolicity • Stochastic processes and SDEs, Nonlinear random walks• Piecewise isometries and tilings• Geometry of foliations• Applications in medicine, biology, neuroscience

http://www.opwdd.ny.gov/institute-for-basic-research/home• Mathematical modeling and data analysis for blood vessels' growth in human placenta during the gestation period• Created the protocol for 3D data collection: 3D surface scanning and micro-CT scans of the specimen• Created a software solution for analysis and visualization of big 3D volumes on BlueHive, a supercomputer at University of Rochester

Conducted research in:• Dynamical systems, attractors, partial and non-uniform hyperbolicity• Stochastic processes and SDEs. Nonlinear random walks• Piecewise isometries and tilings• Applications in medicine, biology, neuroscience

Conducted research in:• Dynamical systems, attractors, partial and non-uniform hyperbolicity• Stochastic processes and SDEs. Nonlinear random walks• Piecewise isometries and tilings

• Developed software and ML algorithms for 3D laser scanners, including• Human face recognition• 3D surface reconstruction from depth maps• Face & facial features detection in 2D and 3D images• Designed and implemented statistical test procedures for new recognition algorithms

• Mathematical research for OCR of hand-written digits• Implemented various SOTA computer vision algorithms in C++, using Artificial Neural Networks, Learning Vector Quantization, Support Vector Machines, Hidden Markov Models, boosting methods