Naama D.

Bio

Software engineering graduate with a specialization in artificial intelligence. Self-taught with natural curiosity and a strong desire for challenges, committed to work and dedicated to achieving goals. Ability to work under pressure, excellent interpersonal skills, excellent at working in a team.

Skills

Python

Docker

CUDA

PyTorch

vLLM

KV Cache

NVIDIA Nsight

HBM

DRAM

Transformer Architecture

Bootcamp Project

SwapKV

High-performance key-value store with memory swapping capabilities

Mentored by: Pliops

Embedded Systems Bootcamp 2025 (Embedded)

Responsibilities:

Studying Transformers and LLM internals
Deep exploration of VLLM architecture
Determining the correct eviction point
mplementing the KV Swap connector
Adding a DRAM eviction mechanism

Additional Projects

End-to-end programming exercise creation platform, including problem statements, solutions, and automated testing for learning environments and technical screening.

• Design and implementation of a RAG pipeline supporting code question generation, including a built-in field extraction data processing tool and a Docker-based ChromaDB vector repository. • Development of Python-based AI components for creating exercises with LLMs. • Execution of prompt engineering benchmarks and optimization of SmolLM2, achieving improved results using 5 retrieved context examples.

Final Project: Nutrition Tracking Website (Full Stack Project)

Nutrition system with a Least Squares algorithm for meal matching based on C# and SQL Server with REST API, combined with Entity Framework for DB management, and integration with an external API.

Course Management System

Course Management system in HTML, CSS, JavaScript and Node.js with MongoDB for managing courses, lecturers and registrations. Emphasis was placed on information security and efficient architecture.

English Level

Fluent