Skip to content
View cristina-v-melnic's full-sized avatar
4 followers · 1 following

Achievements

Achievement: Pull Sharkx2Achievement: YOLOAchievement: Quickdraw

Achievements

Achievement: Pull Sharkx2Achievement: YOLOAchievement: Quickdraw

Block or report cristina-v-melnic

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
cristina-v-melnic/README.md

Welcome!

  • 👋 Hi there, I'm Cristina @cristina-v-melnic - Machine Learning Engineer & Computational Scientist
  • 🔍 Currently seeking new opportunities in ML/AI after completing my role at VION AI developing computer vision solutions for airline systems
  • 🎯 Specialized in translating scientific principles into practical ML applications across industries (manufacturing, healthcare, aerospace)
  • 🧠 Background spanning computational physics, neuroscience, and fluid dynamics with 2+ years in production ML
  • ✨ Passionate about cross-disciplinary problem-solving and building elegant, optimized solutions from first principles
  • 📫 Let's connect: cristina.v.melnic@gmail.com | Portfolio

Featured Projects Portfolio

Here's a glimpse into my work bridging science, AI, and real-world applications:

🤖 Production ML & Computer Vision

  • 🎯 VION AI Passenger Counter - Real-time object detection & tracking system achieving 98-99% accuracy for airline boarding analytics (Feb-Jul 2025)
  • 🔧 Industrial Anomaly Detection - Computer vision system for bolt manufacturing QA, improving accuracy from 95% to 97-98% (prevented 2K+ defective items daily)

🧬 Research & Scientific Computing

  • 🧠 Neural Orientation Selectivity - Computational neuroscience research challenging core assumptions about visual cortex connectivity (ISTA fellowship, 3% acceptance rate)
  • 🌊 Advanced Fluid Dynamics - H(div)-conforming finite element methods for Navier-Stokes equations with perfect mass conservation
  • 🔬 Bayesian Force Inference - Learning physical force fields from data using advanced statistical methods

🎨 Creative AI Applications

  • 💃 Croop.Dance - Interactive choreography learning platform with custom video segmentation and looping tools for dance education
  • 💃 DanceBits - Multimodal AI choreography learning tool using temporal CNNs (preprocessing | API | blog post)
  • 🐅 Animal Pattern Formation - Reaction-diffusion models reproducing realistic fur patterns (giraffe, leopard) with advanced numerical methods

Tech Stack: Python • PyTorch • Computer Vision (YOLO, OpenCV) • Recommender Systems • Scientific Computing • Analytics • ETL • AWS • Git

"Building bridges between fundamental science and practical AI solutions"

Popular repositories Loading

  1. pattern-formation pattern-formation Public

    Modelling pattern formation by numerical integration of systems of coupled convection-diffusion differential equations.

    Python 2

  2. stokes-dg-figures stokes-dg-figures Public

    Analysis of simulation data obtained with the ParMooN finite element package.

    Python

  3. stokes-dg-experiments stokes-dg-experiments Public

    Script that runs experiments on the implemented H(div)-conforming elements for Stokes equations in the finite element package ParMooN.

    C

  4. neural-orientation-tuning neural-orientation-tuning Public

    Model of a single orientation-selective neuron receiving inputs from afferents with different preferred orientations. The project shows what properties the postsynaptic neuron has and how they depe…

    Jupyter Notebook

  5. cristina-v-melnic cristina-v-melnic Public

    Config files for my GitHub profile.

  6. force-inference force-inference Public

    Application to find parameters of inter-particle forces from stochastic trajectories using Bayesian inference.

    Jupyter Notebook