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
Data Science with 2+ years in data-driven solutions & full-stack .NET development, skilled in Python
Aspiring Data Analyst currently pursuing a Master's in Data Science, backed by 2 years of full-stack .NET development experience. Proficient in Python, SQL, and data visualisation tools, with hands-on expertise in building data-driven solutions and working with real-world datasets. Combines solid software engineering principles with advanced analytical skills to deliver scalable, insight-led outcomes. Certified in Microsoft Azure Data Fundamentals (DP-900), with a keen interest in cloud-based data ecosystems.
Kingston University
Master of Science (MSc) · Data Science
August 1, 2025 – June 30, 2027
Sanskrithi School of Engineering
Bachelor of Technology · Computer Science
August 1, 2019 – June 30, 2023
CGI
Associate Software Engineer
July 1, 2023 – July 1, 2025
Bengaluru, Karnataka, India
Real-Time Deepfake Detection using Multi-Modal Deep Learning
September 1, 2025 – Present
Designing a multi-modal deep learning system to detect deepfake video content in near real-time by analysing visual, audio, temporal, and audio-visual synchronisation features. Integrating EfficientNet-B4 for facial artefact detection, Wav2Vec 2.0 for speech inconsistency analysis, and a Video Swin Transformer for temporal coherence modelling. Incorporating Explainable AI techniques (Grad-CAM & SHAP) to provide interpretable detection reasoning, supporting forensic and journalistic use cases. Evaluating on benchmark datasets — FaceForensics++, DFDC, and Celeb-DF — targeting ≥85% accuracy and ≤300ms inference latency per video segment. Addressing critical research gaps: poor generalisation of uni-modal systems, lack of model interpretability, and absence of real-time throughput evaluation.
Train Punctuality & Weather Data Warehouse
January 1, 2024 – December 31, 2024
Designed and implemented a star schema data warehouse integrating railway punctuality and daily weather datasets to support BI reporting and analytical decision-making. Built and executed ETL pipelines to extract, cleanse, and transform raw datasets — standardising formats, resolving missing values, and removing duplicates to ensure data integrity. Modelled shared dimensions (date, station/location) to enable cross-dataset analysis of weather impact on train delays and cancellations. Developed Tableau dashboards visualising monthly punctuality KPIs, delay trends by route, and weather-correlated disruption patterns. Demonstrated how data warehouse design supports proactive operational planning in railway management.
Gesture Phase Classification using Classical Machine Learning
January 1, 2024 – December 31, 2024
Evaluated 5 machine learning classifiers (Decision Tree, Random Forest, KNN, SVM, HistGradient Boosting) on a multi-class gesture dataset with class imbalance. Applied GridSearchCV with stratified 3-fold cross-validation to optimise hyperparameters across all models, using balanced accuracy as the primary evaluation metric. Implemented PCA and Kernel PCA for dimensionality reduction and assessed their impact on classification performance. HistGradient Boosting achieved the best results with 0.623 balanced accuracy and 0.902 macro ROC AUC, outperforming all other classifiers. Concluded that ensemble-based approaches consistently outperform single-model classifiers on imbalanced datasets.
X-ray Image Enhancement Algorithm for Airport Security
January 1, 2023 – May 1, 2023
Designed and built an end-to-end ML-based image processing pipeline to enhance X-ray scan clarity for improved threat detection in airport security systems. Applied advanced computer vision techniques — including edge detection, morphological operations, and contrast enhancement — to isolate and highlight suspicious objects. Engineered data preprocessing steps covering noise reduction, feature extraction, and image normalisation across a dataset of 500+ real-world X-ray images. Achieved a 25% improvement in object detection accuracy, validated using precision, recall, and F1-score metrics. Demonstrated real-world applicability by simulating risk assessment scenarios, showcasing potential for deployment in high-security environments.
Microsoft Azure Data Fundamentals (DP-900)
Microsoft
January 1, 2024 – Present
Data Warehouse Fundamentals
Unknown
January 1, 2024 – Present
Cultural Fit Analysis
The candidate's diverse academic projects, ranging from deepfake detection to airport security and railway punctuality, demonstrate a broad interest in applying data science across various domains. The combination of software engineering experience with a strong academic background in data science indicates adaptability and a willingness to bridge different technical disciplines. The pursuit of a Master's degree while having professional experience shows a proactive and growth-oriented mindset, which aligns well with a culture of continuous learning and innovation. The candidate's ability to work with both structured data (SQL Server) and advanced ML/DL models suggests versatility.
Soft Skills & Operational Fit
The candidate's project descriptions highlight collaboration (e.g., 'Collaborated cross-functionally with business and technical teams') and problem-solving ('Diagnosed and resolved data-related production issues'). The academic projects demonstrate initiative and the ability to work independently on complex problems. The focus on real-time systems and performance metrics (e.g., '≤300ms inference latency', '25% improvement in object detection accuracy') suggests an operational mindset and attention to detail. The candidate's experience in building automated data workflows indicates an understanding of efficiency and scalability.