Job Summary :

We are looking for a AI Scientist to design, develop, and deploy advanced AI models across vision, biometrics, LLMs, and edge/cloud environments. You will contribute to research, build production-ready models, and collaborate closely with engineering teams to deliver high-performance AI systems.

Responsibilities :

AI Model Development :

- Develop state-of-the-art models in computer vision, deep learning, representation learning, and multimodal fusion.

- Implement and fine-tune CNNs, transformers (ViT, Swin, SAM), generative models, and multi-task architectures.

- Design training pipelines, hyperparameter tuning strategies, and evaluation frameworks.

Research & Prototyping :

- Explore novel AI approaches for spoof detection, recognition, cross-domain adaptation, domain generalization, and anomaly detection.

- Build prototypes and perform feasibility studies for next-gen AI products.

Scaling & Optimization :

- Optimize models for high throughput and low latency using quantization, pruning, TensorRT, ONNX Runtime, OpenVINO, and mixed precision.

- Work with GPU clusters for distributed training and experimentation.

Production Deployment :

- Collaborate with platforms and engineering teams to deploy models into cloud/edge production systems.

- Integrate models with pipelines, APIs, microservices, and MLOps workflows.

Documentation & Collaboration :

- Document model architectures, research findings, experiments, and technical specifications.

- Contribute to patents, publications, and internal knowledge-building initiatives.

Qualifications :

- 3 to 6 years of experience in deep learning, computer vision, or AI research.

- Strong experience in PyTorch/TensorFlow, transformers, generative models, and vision architectures.

- Familiarity with ONNX, TensorRT, OpenVINO, and hardware-aware optimization.

- Experience deploying models in real-world applications (cloud or edge).

Preferred Skills :

- Experience in biometrics, recognition systems, and spoof detection pipelines.

- Exposure to federated learning, self-supervised learning, and reinforcement learning.

- Knowledge of MLOps (Docker, Kubernetes, CI/CD).

- Experience working with hardware acceleration platforms (GPU/CPU/MCU/NPU).