Job Description

Overview:

At Rapifuzz, we're dedicated to our mission of 'making security simple,' and empowering organizations against the ever-evolving landscape of digital threats. Our core motivation revolves around securing digital environments and safeguarding sensitive data. Upholding values of integrity, innovation, collaboration, and customer-centricity, we strive to offer unparalleled cybersecurity solutions tailored to meet the unique needs of our clients.Who We Are? As an innovator in the cybersecurity domain, we take pride in our diverse portfolio of next-gen cybersecurity products and services designed to tackle a wide array of security challenges. Our team comprises seasoned cybersecurity professionals with extensive industry experience and deep domain knowledge.

About the role:

We are looking for strong AI/ML Engineers who can build end-to-end deepfake detection capabilities across

image, video, audio, and multimodal data

. You will work on preprocessing, model development, feature fusion, evaluation, explainability, and integration with the backend pipeline. You must be comfortable owning the full ML lifecycle--from data ingestion to baseline model delivery to production-ready scoring.

Job Duties:

Model Development & Multimodal Research

Build and train deep learning models for detecting manipulated media (image/video/audio) using CNNs, Transformers, and multimodal architectures. Implement baseline detectors for: Deepfake image/video detection Lip-sync mismatch detection Face Swap Detection Spatio-Temporal Analysis Illumination and Shading Analysis Voice cloning detection / audio anomaly detection Develop multimodal fusion pipelines (simple concatenation ? classifier).

Preprocessing & Data Engineering

Implement preprocessing for image, video, and audio (frame extraction, spectrograms, MFCCs, lip landmarks). Handle multimodal alignment (audio-video synchronization).

Productionization

Collaborate with backend engineers to expose ML models through APIs. Optimize models for: Real-time inference Scalability Security & adversarial robustness

Required Skills:

Strong Python + PyTorch/TensorFlow. Experience with CNNs, Transformers, GANs, audio ML, or multimodal learning. Experience with image/video/audio preprocessing pipelines. Understanding of digital forensics, media manipulation, or adversarial ML. Experience with model deployment workflows (TorchServe, ONNX, Triton preferred).

Preferred Skills:

Experience with C2PA validation, watermark verification, or media forensics. Experience with real-time video pipelines, FFmpeg, or GPU optimization. Knowledge of scalable ML serving and cloud infrastructure.

Soft Skills:

Ownership-driven mindset; ability to independently deliver ML components. Excellent communication, structured problem solving.
Job Type: Full-time

Pay: ₹12,242.75 - ₹123,363.58 per month

Benefits:

Health insurance
Application Question(s):

How many years of hands-on experience do you have in Deep Learning / Applied ML? Which model architectures have you used in production or research? Which ML frameworks do you use most? What is your current CTC in Lacs per annum? What is your expected CTC in Lacs per annum? What is your notice period?
Location:

Gurgaon City, Haryana (Required)
Work Location: In person