Help & Supportcareers@gravityer.com
Data Science with less than a year in Machine Learning & Time-Series forecasting.
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
Detail-oriented Data Scientist seeking to leverage advanced analytical skills in Machine Learning and Forecasting to deliver measurable value within a fast-paced, innovation-led environment. Proven ability to build end-to-end Machine Learning models, specializing in Time-Series and viewership forecasting. Skilled in feature engineering and data cleaning to transform raw datasets into actionable business insights. Passionate Data professional dedicated to transforming complex unstructured data into actionable business insights through advanced feature engineering and state-of-the-art AI technologies.
Predictive Maintenance for Medical Equipment
June 1, 2026 – Present
Engineered a predictive maintenance framework using Gradient Boosting and Random Forest to estimate the Remaining Useful Life (RUL) of critical medical assets, achieving an R^2 \approx 0.30 and a Mean Absolute Error of 76 days. Developed complex interaction features, such as Age x Usage and Temp \times Volt, to capture non-linear failure triggers, increasing model sensitivity to operational stress beyond chronological age. Implemented Hierarchical Clustering to group 7,000+ medical devices (e.g., Ventilators, MRIs) by failure signatures, identifying that maintenance history accounts for up to 46.8% of total failure risk. Mitigated data leakage by systematically isolating target-correlated variables, ensuring a robust and unbiased model capable of generalizing across diverse hospital sensor data. Extracted actionable business insights that enable a transition from reactive to proactive maintenance, potentially reducing hospital equipment downtime and optimizing CAPEX allocation.
Broadcast Viewership Forecasting
June 1, 2026 – Present
Developed a machine learning forecasting model to predict broadcast viewership up to 15 days ahead using historical audience data. Applied time-based feature engineering and direct multi-step regression techniques while preventing data leakage. Achieved strong predictive performance with an R² score of 0.89 on unseen test data. Recorded low forecasting error with MAE of ~2.3 and RMSE of ~3.7, indicating high model reliability.
Predictive Maintenance for Aircraft Engines
June 1, 2026 – Present
Built an end-to-end predictive maintenance system for aircraft engines using the NASA C-MAPSS time-series dataset with multivariate sensor data. Reframed Remaining Useful Life (RUL) prediction into a 4-class maintenance classification problem (Healthy, Soon, Warning, Critical) to support operational decision-making. Prevented data leakage by performing engine-level train/validation splits, ensuring realistic evaluation on unseen engines. Engineered temporal features (rolling mean, rolling standard deviation, and cycle-to-cycle deltas), expanding the feature set from 24 to 87 features. Trained a Random Forest classifier achieving ~85% validation accuracy, with ~97% recall for Healthy cases and ~88% recall for Critical failures. Enhanced model interpretability by analyzing top sensor importance, highlighting key warning signals used to identify imminent engine failures.
Inview: AI-Driven Bilingual Video Interviewing & Proctoring Platform
June 1, 2026 – Present
Developed an end-to-end AI platform for automated candidate screening using LLMs (Qwen 2.5) and Speech-to-Text (Whisper-v3) for deep semantic analysis of interviews. Engineered a Bilingual NLP Pipeline (Arabic/English) capable of extracting technical proficiency, communication scores, and behavioral traits from unstructured audio data. Implemented a Computer Vision (OpenCV) proctoring system to monitor candidate presence and detect multi-face security alerts during live sessions. Built a professional-grade Glassmorphism Dashboard using Streamlit, featuring dynamic AI-generated reports and a multi-factor scoring matrix for HR decision-making. Automated the deployment pipeline to Streamlit Cloud, integrating complex data workflows with real-time video processing.
IBM Data Science Professional Certificate
Coursera
June 1, 2026 – Present
IBM Machine Learning
Coursera
June 1, 2026 – Present
Data Analysis with Python
Coursera
June 1, 2026 – Present
Deep Learning & Reinforcement Learning
Coursera
June 1, 2026 – Present
SQL Certification
DataCamp
June 1, 2026 – Present
Explore AI & Generative AI
Microsoft Learn
June 1, 2026 – Present
Certification Generative AI Elevate Your Data Science Career
Coursera
June 1, 2026 – Present
Certification Machine Learning with Python
Coursera
June 1, 2026 – Present
Cultural Fit Analysis
The candidate's diverse project portfolio, including predictive maintenance for different domains (aircraft, medical equipment), broadcast viewership forecasting, and an AI-driven interviewing platform, demonstrates a broad interest in applying data science to various real-world problems. The pursuit of numerous certifications also indicates a proactive learning attitude. This diversity and continuous learning align well with an innovative and fast-paced environment. The target role of 'Data Science' is well-aligned with the candidate's demonstrated skills and project focus.
Soft Skills & Operational Fit
The candidate's project descriptions indicate strong analytical and critical thinking skills, along with a business-oriented approach to data interpretation. The 'Inview' project suggests an ability to work on end-to-end solutions and integrate various technologies, which points to good problem-solving and potentially collaborative skills. However, without direct experience or psychometric test results, it's difficult to fully assess stress handling or team collaboration.