Help & Supportcareers@gravityer.com
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
AI Engineer with less than a year in Machine Learning, Computer Vision, and NLP.
Fresh Graduate AI Engineer specializing in machine learning, Computer Vision (CV), Natural Language Processing (NLP), and intelligent automation. Skilled in designing end-to-end AI systems - from data preprocessing and model training through deployment and workflow automation. Experienced building real-world solutions using deep learning, LLM orchestration (LangChain, CrewAI), and browser automation, with a strong focus on practical, scalable, and privacy-conscious architectures.
The British University of Egypt (BUE)
BSc Computer Science · Artificial Intelligence
November 1, 2020 – July 1, 2024
CaptionAI - AI-Powered Multilingual Subtitle Generator
January 1, 2026 – Present
Built a full-stack AI subtitle generation platform that transcribes speech and translates it into 10+ languages automatically from any video or audio file. Integrated faster-whisper large-v3 on GPU (CUDA) for 98%+ accuracy ASR with word-level timestamps and noise filtering; translation powered by Claude Sonnet via Anthropic API for cinematic-quality output. Designed a FastAPI backend with async job pipeline, in-memory job store, and SRT file export; React frontend with real-time progress tracking and per-language subtitle preview and download. Implemented full RTL language support (Arabic) and resolved complex Windows CUDA/cuDNN library compatibility issues for local GPU inference.
MedTech Pulse: Automated Knowledge Extraction
January 1, 2026 – Present
Developing an end-to-end pipeline to convert unstructured technical logs and medical machine reports into a structured, searchable knowledge base. Orchestrating local vision models and LLMs via LangChain, CrewAI, and Ollama for privacy-focused data reasoning; Firebase for real-time data sync.
Multilingual RAG Platform
January 1, 2026 – Present
Built a Retrieval-Augmented Generation (RAG) platform supporting multilingual document ingestion, semantic search, and LLM-powered question answering. Integrated vector store retrieval with LangChain pipelines; designed for multi-language support and scalable document handling.
LinkedIn Job Application Automation (NLP + Web Automation)
January 1, 2026 – Present
Developed an automated LinkedIn Easy Apply system that parses job descriptions using NLP to extract title, experience level, and required skills. Implemented keyword-based CV matching to improve ATS compatibility; automated the full application workflow via browser automation.
Sign Language Recognition System (Graduation Project)
January 1, 2024 – June 1, 2024
Built a real-time CV system supporting ASL, ARSL, and BSL with live webcam inference and text-to-speech output. Models: EfficientNetB0 (ASL), CNN (ARSL), VGG19 (BSL); experimented with SVM + HOG features. Datasets: ASL ~367K images, ARSL ~50K, BSL ~3.8K frames. Responsibilities: data aggregation & augmentation, model training, spelling/autocorrect post-processing, real-time inference demo.
View ProjectAG-News Topic Classification
January 1, 2024 – June 1, 2024
Implemented BERT and DistilBERT transformer classifiers to categorize news into four topics; full pipeline from cleaning and tokenization to evaluation.
Book Genre Prediction (Deep Learning + NLP)
January 1, 2024 – June 1, 2024
Evaluated Word2Vec (CBOW, Skip-Gram) and GloVe embeddings combined with CNN and LSTM architectures for multi-class genre classification. Handled class imbalance and model comparison across 10 genres; 4,658 book records.
Voice / Speaker & Gender Classification
January 1, 2024 – June 1, 2024
Extracted MFCCs and spectral acoustic features from 3,168 speech samples; trained and evaluated classical ML classifiers.
House Price Prediction - Deployed Web App
January 1, 2023 – December 31, 2023
Built and deployed a Flask web application serving a serialized Linear Regression model with real-time inference via web UI. Implemented preprocessing pipeline (missing values, one-hot encoding) and model serialization with Pickle.
Introduction to Machine Learning Training
AITB
August 1, 2023 – Present
Cultural Fit Analysis
The candidate's academic projects show a strong interest and focus on practical AI applications, aligning well with a role that requires building real-world solutions. The diversity of projects (CV, NLP, ASR, RAG, automation) indicates a broad technical curiosity and adaptability. The volunteering experience suggests some engagement in extracurricular activities, which can be a positive indicator for cultural fit, though its direct relevance to a technical role is limited.
Soft Skills & Operational Fit
The candidate's project descriptions indicate a proactive and hands-on approach to problem-solving, particularly in resolving complex technical issues (e.g., Windows CUDA/cuDNN compatibility). The academic projects demonstrate an ability to work independently on complex tasks. However, without specific behavioral assessment data or interview transcripts, it's difficult to fully assess communication, teamwork, and stress handling.