Senior Computer Vision & Deep Learning Scientist — Liveness & Deepfake Detection
Our mission is to create innovative, robust, and user-friendly digital identity solutions. We are looking for a passionate and skilled Computer Vision & Deep Learning AI Scientist specializing in Liveness and Deepfake detection to join our dynamic team. Your work will directly impact the security and reliability of VIDA's identity verification systems.
Responsibilities:
- Design, train, and deploy CV/DL models for presentation attack detection (PAD) and deepfake detection (print/replay, 2D/3D masks, screen replays, injection, face-swap/GAN/diffusion).
- Own datasets end-to-end: collection strategy, labeling specs, quality gates, hard-negative mining, bias/fairness audits.
- Build realistic benchmarks and red-team suites; report metrics (FAR/FRR), ROC/PR, calibration; run ablations and adversarial stress tests.
- Ship models with clear latency/SLA targets; partner with engineers on scalable inference APIs and monitoring (drift, false-positive hot spots).
- Translate fraud patterns into model features and detection rules; collaborate with product, fraud ops, and key customers.
- Track state-of-the-art CV/DL and adopt techniques that improve security, latency, or cost; align with ISO/IEC 30107 (PAD) best practices.
- Work independently or in a team to solve complex problem statements
Requirements:
- Advanced degree in a quantitative field or equivalent experience, and 3+ years building/launching computer vision & deep learning models (biometrics, PAD, or deepfake detection preferred).
- Strong CV/DL fundamentals: modern backbones/transformers, temporal models, contrastive/metric learning, robustness techniques.
- Proficient in Python and PyTorch (or TensorFlow); OpenCV/FFmpeg; proven track record taking models from notebook → production with experiment tracking, CI/CD, and post-deploy monitoring.
- Clear communication, ownership mindset, and bias to ship.
Nice to have:
- Deepfake forensics (GAN/diffusion artifacts, watermark/signature checks).
- On-device/low-latency inference (TensorRT, ONNX Runtime, TFLite, Core ML, NNAPI), CUDA kernels, or inference compilers.
- Familiarity with cloud platforms like AWS, GCP, or Azure for model deployment.
- Familiarity with ISO/IEC 30107 (PAD), ISO/IEC 19795 (biometric performance), ISO/IEC 39794 (biometric data interchange).
- Synthetic data/simulation to cover rare attacks.
About VIDA Digital Identity
- VIDA is a government-licensed certificate authority (CA) operating under Indonesia’s Ministry of Electronics and Information Technology. Our key offerings include:
- Legally valid digital signatures ensuring secure and scalable identity solutions.
- Trusted identity verification services for industries such as BFSI, eCommerce, telecommunications, and healthcare.
- Advanced AI-driven fraud prevention: VIDA leverages sophisticated AI models to detect and prevent deepfake frauds, enhancing security and trust in digital interactions.
- Seamless integration to help businesses reduce onboarding friction while prioritizing user privacy and data security.
- For more information, you may visit our website at https://vida.id