Stepan Lavrinenko
Staff Data Scientist Genai Nlp @Seequent
Signup Ā· Get unlimited contacts
WORK HISTORY
Staff Data Scientist Genai Nlp @Seequent
Calgary, AB, CA
Building custom chatbot interfaces- RAG and Knowledge Graph based RAG- ADK and Langgraph multiagent apps- Research and productizing memory systems/ agent swarms in agentic architectures - Graph ML algorithms- LLM fine-tuning and design- Optimization models- Deep Learning-based inversion modeling
EDUCATION
University of Alberta
Master of Integrated Petroleum Geosciences, Geophysics and Seismology, Geology, Reservoir Engineering, Petrophysics, Petroleum Systems
University of Alberta
Bachelor of Science (BS), Specialization in Geophysics
SKILLS
ABOUT STEPAN LAVRINENKO
As a Staff Scientist at Seequent/Bentley, I leverage over 9 years of experience in Machine Learning (ML) to tackle complex challenges within the mining industry. I worked with geospatial models, speeding up TB-scale spatial joins from days to hours; built custom GenAI apps to help onboard new users to complex geoscience software packages; forecasted and prevented damaging vibrations to engineering structures by 90%; saved $8M/year per minesite via physics-informed neural network blast modeling and a lot more.In my current role, I spend about 50% of my time building GenAI apps with langgraph and the other 50%- building deep neural networks for geoscience and engineering use cases (computer vision, transformers, GNNs, etc).My responsibilities encompass formulating technical problems, designing and executing experiments, building and deploying prototypes, and implementing MLOps pipelines to ensure quality and scalability. Additionally, I engage in the hands-on work of a Senior Data Scientist, encompassing the authoring and review of processing code, as well as the setup and tuning of ML model and neural network pipelines. I increased the code quality in my team 3x by an independent review process and enforced the usage of code patterns for reusability. I really enjoy learning new things and connecting with people across a range of industries, so don\'t hesitate to reach out if you\'d like to get in touch:Telegram:@ukr_can_dsEmail: sā¢ā¢ā¢ā¢ā¢ā¢ā¢@ualberta.caCompetencies: Data Science, Machine Learning, Data Engineering, Deep Learning, Artificial Intelligence, Computer Science,programming, Python, SQL, pyspark, mojo, pandas, numpy, oop,team management, leadership, mentorship,statistics, statsmodels, A/B testing, PyMC, causal models, DoWhy, Causalinference, propensity score matching,markov networks, bayesian networks, pomegranate,scikit-learn, hyperopt, optuna,Keras, tensorflow, pytorch, Neural Networks, Computer vision, transformers, Siamese Networks,Time series, LSTM, RNN, scipy, signal processing, huggingface,langchainn, LLM, RAG, LLM fine-tuning, NLP, spacy, nltk, genism,mosaic, h3, geospatial, PINN (physics-informed neural networks), Physics-based deep learning, applied math, geophysics, seismics, geology,clusterization, dimensionality reduction,survival modeling, survival analysis, scikit-surv,Data Analytics, Data storytelling, Data visualization, matplotlib, seaborn, plotly, data analytics,Azure, AWS, vertex AI, databricks,
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.