Natanel Richey

AI/ML Developer | Full-Stack Developer

Software-developer skilled in Python and JavaScript focused on Applied Machine Learning and Web/App Design. Experience in training AI models, building full-stack websites, data analysis (AI enhanced). Methodical and analytical, hard-working and efficient, fast-learning, hungry for excellence and passionate about bridging the gap between human and machine intelligence.

LinkedIn GitHub

✉️ natanaelrichey.work@gmail.com 📱 0584171094

Professional Experience

Nov 2025 - Present

Machine Learning Engineer

BabyClue

📍 Part-time

Built and evaluated a multi-class infant-cry classification pipeline, focusing on robust validation, interpretability, and reproducibility.
Developed end-to-end ML evaluation workflows in Python
Ran rigorous, comparative model testing using nested CV with fold-level and aggregated analysis (macro/micro F1, per-class precision/recall/F1, confusion matrices, latency) to reliably determine which approaches generalized best across splits
Built the full training pipeline combining audio preprocessing (spectrogram-based features / AST embeddings), data augmentation, feature construction (late-fusion of embedding + acoustic features), and model training/inference across multiple model families (e.g., MLP, LightGBM, plus classical baselines like SVM/Nearest Centroid where applicable)

Technologies:

PythonMachine LearningData-Centric AIHyperparameter Optimization

Education

2021 - 2025

BSc. in Computer Science, B.A. in Psychology (Hons.)

Hebrew University of Jerusalem

📍 Jerusalem, Israel

Languages:

PythonCC++Java

Skills:

NLP and Machine LearningOOPOperating SystemsAlgorithms and Data StructuresStatistics and Computational MethodsData Science

Portfolio

A selection of projects showcasing my technical skills and self-learning abilities

Demo Available on Request

TheraBot - Fine-tuned AI Model

Fine-tuned Llama model using HuggingFace Transformers with LoRA ranking on therapy conversation datasets.

PythonAI Model TrainingLoRA Fine-tuningNLPRAG implementation

Key Highlights:

✓RAG implementation using FAISS vector store with E5 embeddings
✓Weights & Biases experiment tracking and model monitoring

GitHub

Demo Available on Request

MealCreator - Full-Stack React Website

A full-stack Progressive Website for meal planning and pantry management.

Next.jsTypeScriptMongoDBReact QueryCloudinaryVercelPWA

Key Highlights:

✓Real time data sync using React Query with optimistic updates and intelligent cache invalidation, reducing API calls by 80%+
✓TypeScript implementation across Next.js components, RESTful backend API, MongoDB and Mongoose schemas

GitHub