dsvolk Goto Github PK

Full Stack Data Science Engineer @ Toptal, PhD in Mathematics

I am a top-notch full-stack data scientist and ML engineer, highly skilled in modern generative AI, machine learning, MLOps, data analysis, mathematical modeling, data pipelines, big data, and cloud infrastructure. My expertise includes LLMs and AI chatbots, time series analysis and forecasting, geospatial data analysis, natural language processing, probabilistic modeling, data engineering, and team leading.

I am Ph.D. in mathematics, and I did research in Math, Physics, Neuroscience, and Medicine.

2023 - PRESENT

• Created OkGPT, an AI personal assistant messenger bot. 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.
• Integrated multiple APIs, including Telegram, OpenAI, Cohere, Anthropic, Google, Redis, Amplitude, Datadog, and more.
• Supervised three team members and a few external collaborators.

2024

• Trained a custom BERT deep neural network to identify AI-generated software code.
• Created a data pipeline to prepare model training data from open source code repositories on GitHub.
• Architected a test suite for classification models, including validation datasets and testing procedures.
• Deployed the classification models to production and containerized microservices in AWS cloud.

2023

• Developed an MVP of an app to query enterprise data in natural language. Given access to a database and a question in natural language about the data, the app would output the answer as a plot or a small table.
• Engineered and fine-tuned the prompts to improve the quality and correctness of SQL code generation.
• Created an automatic annotator for the database columns and the final table.

2023

• Created a machine learning model to automatically detect malicious smart contracts before they can cause harm.
• Built a visualization tool for model output to audit its decisions.
• Set up automatic model deployment to AWS cloud platform as a Lambda serverless function.

2021 - 2022

• Architected and directed the creation of a core Similarity engine to score candidates.
• Used pre-trained NLP deep neural networks to create semantic text embeddings, which significantly increased the Similarity engine output results.
• Created a Big Data pipeline in Databricks and Spark to enrich the input data and prepare the features for ML.
• Prepared custom deep learning models to build richer embeddings, including various data sources and metadata.

2020 - 2021

• Built a foundational end-to-end machine learning solution that predicts fair prices of real-estate properties, thus eliminating a need for manual assessment and enabling the company to run its business by providing quick responses to its customers.
• Designed and implemented an automatically refreshing ETL pipeline that injects, cleans, joins, and enriches new data from AWS S3 storage daily.
• Developed an interpretable machine learning model with Scikit-learn, CatBoost, Lifelines, FBProphet, FAISS, and SHAP that consists of several submodels and satisfies business monotonicity constraints.
• Supervised other data science team members and coordinated with the engineering team.

2017 - 2019

• Created a machine learning model that predicted revenues for a retail store chain based on store location, local demographic data, GIS features, seasonality, and other factors.
• Developed and deployed an interpretable machine learning model that scored B2B customers for payment default risks and provided explanations for the scores. The model massively reduced workload for weekly risks assessment.
• Built a probabilistic Bayesian machine learning model to predict which apartment buildings still under construction would fail to be commissioned in time. The model helped reduce the funds needed to hedge risks by two times.
• Developed and deployed NLP models to automatically label a vast body of housing contracts by contract type and extract contractor party names, address entities, and other attributes.

2004 - 2007

• Designed and implemented biometric machine learning face recognition algorithms.
• Created and implemented statistical test procedures for new recognition algorithms.
• Developed calibration procedures from 3D laser and flash scanners.

2017 - 2020

2015 - 2017

• Invented a novel mathematical method for cross-frequency synchronization analysis in the human brain.
• Implemented the method as a MATLAB toolbox and ran tests confirming that the results agreed with previously known scientific data.
• Prepared and published the method and findings in a top-level journal.

2013 - 2015

• Discovered a new geometric phenomenon accountable for the rigidity of certain mathematical models related to heat conduction in crystals.
• Discovered a new stability property of attractors of multidimensional piecewise isometry maps related to Markov field models.

2012 - 2013

• Established that the rotation numbers of circle maps' semigroups define their generators.
• Discovered a fractal structure of attractors of piecewise isometry maps related to Markov field models.
• Lectured a PhD-level course on the structural stability of dynamical systems.

2010 - 2012

• Discovered a new class of dynamical systems that have persistent massive attractors.
• Established a deep relationship between skew product dynamical systems over Markov chains and nonlinear random walks.

• An accidental image feature that appears but not disappears (with T. Sawada) // Journal of Mathematical Psychology, 119, 2024
• Generalized Cross-Frequency Decomposition: A Method for the Extraction of Neuronal Components Coupled at Different Frequencies (with I. Dubinin, A. Myasnikova, B. Gutkin, V. Nikulin) // Frontiers in Neuroinformatics, 2018

• Fast-slow partially hyperbolic systems versus Freidlin-Wentzell random systems (with J. De Simoi, C. Liverani, C. Poquet) // Journal of Statistical Physics, 166:650--679, 2017 [see also arXiv:1607.04319]
• Nonwandering sets of interval skew products (with V. Kleptsyn) // Nonlinearity, 27:1595--1601, 2014 [see also arXiv:1302.6929]
• Physical measures for nonlinear random walks on interval (with V. Kleptsyn) // Moscow Mathematical Journal, 14(2*):339--365, 2014 [see also arXiv:1110.2117]
• Translation numbers define generators of $\mathbb{F}^{+}_{k} → \mathrm{Homeo}_{+}(S^1)$ (with T. Golenishcheva-Kutuzova, A. Gorodetski, V. Kleptsyn) // Moscow Mathematical Journal, 14(2):291--308, 2014 [see also arXiv:1306.5522]
• Almost every Interval Translation Map of three intervals is finite type // Discrete and Continuous Dynamical Systems - Series A, 34(5):2307--2314, 2014 [see also arXiv:1203.3405]
• Persistent massive attractors of smooth maps // Ergodic Theory and Dynamical Systems, 34(2):693--704, 2014 [see also arXiv:1108.5330]
• Cascades of $\varepsilon$-invisibility (with Yu. Ilyashenko) // Journal of Fixed Point Theory and Applications, 7(1):161--188, 2010 [see also arXiv:0906.3567]
• Theorem on the density of separatrix connections for polynomial foliations in $\mathbb{C}P^2$ // Journal of Mathematical Sciences, 150(5):2326--2334, 2008
• The density of separatrix connections in the space of polynomial foliations in $\mathbb{C}P^2$ // Proceedings of the Steklov Institute of Mathematics, 254(1):169--179, 2006

• Attractors of Piecewise Translation Maps // arxiv:1708.03780

• Mapping placental topology from 3D scans, the graphic display of variation in arborisation across gestation (with C. Salafia, M. Yampolsky, C. Stodgell, P. Katzman, J. Culhane, P. Landrigan, S. Szabo, N. Thieux, J. Swanson, N. Dole, M. Varner, J. Moye, R. Miller) // Placenta, 34(9):A73--A74, 2013 (see also Erratum)

• Fast-slow partially hyperbolic systems and pathological foliations (with De Simoi, C. Liverani, C. Poquet) // International Conference "Anosov systems and modern dynamics", 2016, pp. 25–27, ISBN 978-5-98419-073-2
• Dynamics of Piecewise Translations // Proceedings of International Conference "Dynamics, Bifurcations and Strange Attractors", 2013, pp. 112–113
• Interval Translation Maps of Three Intervals // Proceedings of International Conference on Differential Equations and Dynamical Systems, 2012, ISBN 978-5-98419-046-6
• Persistent massive attractors of smooth maps // Proceedings of International Mathematical Conference “50 Years of IITP”, 2011, ISBN 978-5-901158-15-9
• Skew products with interval fiber // Conference on Geometry and Topology of Foliations, 2010, p. 23
• Thin attractors (with V. Kleptsyn) // Topology, Geometry and Dynamics: Rokhlin Memorial, 2010, pp. 70–72
• The density of separatrix connections in $\mathbb{C}^2$ // International conference “Differential equations and related topics” dedicated to I. G. Petrovskii, 2004. Book of Abstracts, p. 241 (Russian).

• ERC Advanced Investigator Grant MALADY 246953 [European Union]
• “Young SISSA Scientists” (Principal Investigator) [Italy]
• CNRS 10-01-93115 [France]
• PRIN [Italy]
• CNRS 05-01-02801-CNRS_a [France]
• CRDF RM1-2358 [USA]