Debjyoti Ghosh
AI/ML Research Engineer
Undergraduate researcher and AI/ML engineer specializing in AI Security, Trustworthy AI, Multimodal Learning, and Media Forensics. Experienced in deepfake detection, context-aware NLP, voice biometrics, and scalable backend infrastructure.
Scroll to explore
Skills & Technologies
Tools, frameworks, and programming languages I use to build research models and production backend infrastructure.
My Research Work
Cutting-edge research in AI/ML with focus on deepfake detection, forensics, and biometric applications
Multimodal Deepfake Detection via Audio-Visual Fusion
95% (0.97 AUC) Accuracy
Architected a scalable multimodal deepfake detection pipeline utilizing a Transformer-based attention mechanism to fuse spatial, temporal, and acoustic embeddings, achieving 95% accuracy and 0.97 AUC.
Key Highlights
- Integrated ResNet-50 visual encoder, BiLSTM temporal encoder, and 1D-CNN audio encoder processing Mel-spectrograms.
- Implemented custom synchronization loss function to enforce phoneme-viseme alignment verification.
- Evaluated on FakeAVCeleb benchmark dataset demonstrating superior cross-modal robustness.
Technologies
Context-Aware Hate Speech Detection via Post-Comment Joint Modeling
95% (0.9911 AUC) Accuracy
Created a context-aware moderation framework replacing isolated text analysis with post-comment joint modeling, achieving 95% accuracy and ROC-AUC of 0.9911.
Key Highlights
- Structured four-class reaction taxonomy mapping discourse to explicit moderation actions.
- Fused parallel features including SBERT semantic similarity, emotion distributions, and DistilBERT probabilities.
- Validated framework robustness on curated social media interaction dataset.
Technologies
Edge-Based Voice Biometric Verification & Cloud Anti-Spoofing
97.5% Accuracy
Constructed a speaker recognition system combining edge-based audio acquisition with a centralized cloud backend, achieving 97.5% authentication accuracy in controlled environments.
Key Highlights
- Formulated embedding pipeline utilizing MFCCs and spectral features via Gaussian Naive Bayes classifier.
- Engineered FastAPI backend and web interface facilitating secure user enrollment and parallelized inference.
- Evaluated against environmental noise and spoofing vulnerabilities.
Technologies
Developer Extensions & Tools
Open-source tools, IDE extensions, and developer utilities published for the global community.
AntiGravity (Dev Timekeeper) -- VS Code Extension
- Created and published a developer productivity extension for VS Code with over 1,300 downloads on Open VSX Registry.
Professional Journey
Backend Engineer (Python Developer)
Nov 2025 -- Feb 2026ThreatPurge Private Limited
Engineered scalable backend services and asynchronous ETL pipelines for phishing detection and brand abuse monitoring.
Full-Stack AI Engineer (Freelance)
Mar 2026 -- PresentElaina Legal AI
Architected dual-facing Agentic AI legal assistant and Bayesian Neural Network (BNN) for legal outcome prediction.
Awards & Recognition
Hackathons, robotics competitions, publications, and certificates
IEEE INDICON 2025 Paper Recognition
IEEE INDICON 2025 Paper Recognition
Publication PDF
CSI CreaXion Project Competition – 2nd Place
CSI CreaXion Project Competition – 2nd Place
Hackathon
IEM HackOasis 2.0 – 2nd Place
IEM HackOasis 2.0 – 2nd Place
Hackathon
Srijan'25 Homecoming – 1st Place
Srijan'25 Homecoming – 1st Place
Event
URECKON Robo Sumo – 1st Place
URECKON Robo Sumo – 1st Place
Robotics
URECKON Direction – 1st Place
URECKON Direction – 1st Place
Robotics
URECKON Robo League – 2nd Place
URECKON Robo League – 2nd Place
Robotics
URECKON Robo Dangal – 2nd Place
URECKON Robo Dangal – 2nd Place
Robotics
Amazon ML Summer School 2025 – Selected Participant
Amazon ML Summer School 2025 – Selected Participant
Certificates PDF
Backend & AI/ML Engineer Intern – ThreatPurge Pvt. Ltd.
Backend & AI/ML Engineer Intern – ThreatPurge Pvt. Ltd.
Certificates PDF
AI for Forensics and Biometric Applications (AIFB 2025)
AI for Forensics and Biometric Applications (AIFB 2025)
Publication PDF
Get in Touch
I'm open to freelance projects, AI/ML engineering consultations, research opportunities, and technical collaborations.