Open to opportunities

Valerii Kvan

@valeriikvan

Message

I build performance-critical C++/ML systems for CAD geometry and production pipelines.

Georgia

Message

What I'm looking for

I’m looking to ship production-grade, modular C++/ML systems—especially CAD geometry, reconstruction, and optimisation—within an Agile team that values performance, clean design, and measurable product impact.

I’m a C++/ML engineer with 5+ years building performance-critical algorithmic systems—computational geometry, BRep/mesh processing, and optimisation in production CAD software. I also ship ML pipelines in production, using PyTorch and Scikit-learn alongside custom statistical models and nonlinear optimisation. I’m drawn to turning ambiguous research problems into modular code that real customers can rely on.

At Huawei (Remote), I implement real-time geometry editing algorithms in C++ for direct modeling capabilities. I developed a mesh-to-surface reconstruction algorithm (C++, ACIS/OCCT, BRep/Mesh) to restore BRep models from triangulated meshes and reduce manual re-modeling effort. I’m also integrating a RAG LLM agent with an IDE to automate code navigation and documentation lookup, speeding up feature delivery.

Previously at BIOCAD, I designed an analytical dose-response pipeline that shortened drug research timelines before market release. I built non-linear mixed effect (NLME) and survival models for virtual patient simulation and missing data imputation, reducing required sample sizes by up to 30–40% and lowering recruitment costs. I engineered reusable preprocessing for batch effect detection, ran A/B testing and comparative statistics to support regulatory approval, and accelerated R&D decisions with LLM-based literature retrieval and synthesis.

Earlier, I delivered predictive ML models that improved cardiovascular risk forecasting beyond standard clinical scales (about 10 percentage points in ROC AUC) and supported peer-reviewed publications. I also developed BRep geometry processing algorithms at CADEX with meaningful performance gains for SaaS CAD, and built BIM automation using the Revit API to reduce engineering drafting time. My Master’s training in applied mathematics (with honours) strengthens my foundation in linear algebra, numerical methods, and complexity analysis—so I can engineer solutions that stay fast and correct under real constraints.

Experience

Work history, roles, and key accomplishments

Current

Software Engineer (C++/ML)

Current

Huawei

Oct 2025 - Present (8 months)

Implemented high-performance real-time geometry editing algorithms to enable faster early-stage CAD design. Developed mesh-to-surface reconstruction to restore BRep models from triangulated meshes and integrated a LangChain RAG agent into an IDE to automate documentation/code navigation.

C#BRep Mesh Processing Computational Geometry VTK Performance Optimization

Data Scientist

BioCAD

Jun 2024 - Oct 2025 (1 year 4 months)

Built analytics pipelines for dose-response analysis and automated scientific literature retrieval to accelerate R&D decision-making. Developed non-linear mixed effect and survival models for virtual patient simulation and missing data imputation, reducing required sample sizes for analysis by 30–40% and lowering recruitment costs.

Python scikit learn NLME (Nlmixr)Survival Analysis data preprocessing A B Testing LLM Prompt Engineering

Data Scientist

Almazov Centre

Jun 2024 - Dec 2024 (6 months)

Created feature-importance and statistical analysis pipelines to identify key predictors for health outcomes and support preventive therapy decisions. Evaluated added predictive value of abdominal obesity for cardiovascular events using AdaBoost and published results in a peer-reviewed European medical journal.

Python AdaBoost scikit learn Scipy Statistical Analysis Feature Importance A B Testing

Data Scientist

Almazov Centre

Nov 2023 - Jun 2024 (7 months)

Engineered predictive ML models (Random Forest, XGBoost, PyTorch) to forecast cardiovascular risk, achieving ~10 percentage points better ROC AUC than standard clinical scales. Built ML pipelines to identify restenosis and plaque microbiome predictors in carotid endarterectomy patients, contributing to peer-reviewed publications.

Python Random Forest XGBoost PyTorch scikit learn ML Pipelines Statistical Analysis Feature Importance

Software Engineer

CAD Exchanger

Mar 2023 - Nov 2023 (8 months)

Developed novel BRep geometry processing algorithms to improve manufacturing cost estimation accuracy for a SaaS CAD system. Optimized feature recognition with Open CASCADE to deliver 30% performance gains for critical processing modules.

C#Open CASCADE NURBS BRep Geometry Processing Performance Optimization Computational Geometry

Software Engineer (BIM)

Industrial Systems for Streaming Data

Mar 2021 - Dec 2022 (1 year 9 months)

Designed algorithms for automated 3D-to-2D projection of building models, reducing engineering drafting time by 20%. Developed a BIM system using the Revit API to automate design workflows and improve team productivity in an Agile environment.

C#Revit API BIM Workflows Agile Scrum Performance Optimization

Education

Degrees, certifications, and relevant coursework

ITMO University

Master’s Degree (with honours), Bioinformatics and System Biology

2021 - 2023

Master’s degree (with honours) in Bioinformatics and System Biology at ITMO University from 2021 to 2023.

Peter the Great St.Petersburg Polytechnic University

Bachelor’s Degree, Computer Science & Engineering

2017 - 2021

Bachelor’s degree in Computer Science & Engineering (2017–2021) with a thesis on a Computer Vision module (ROS2, Gazebo, C++, YOLO) for autonomous vehicle SLAM enhancement.