Biocube - Deep Learning Engineer - PyTorch/Tensorflow

Key Responsibilities :

- Design, train, and optimize deep learning models for computer vision and biometric recognition tasks including face identification, speaker identification, fingerprint and palmprint recognition, and eye socket-based identification.

- Apply model compression techniques like quantization, pruning, and knowledge distillation to optimize inference for deployment on edge devices.

- Work with lightweight model architectures such as MobileNet, and deploy models using ONNX, TensorFlow Lite (TFLite), and CoreML.

- Develop and test liveness detection mechanisms to enhance security and robustness.

- Collaborate with cross-functional teams to integrate models into production pipelines.

- Automate data preprocessing, annotation, and augmentation workflows.

- Maintain thorough documentation of experiments, code, and deployment strategies.

Requirements and Skills :

- Solid understanding of deep learning fundamentals and architectures, particularly CNNs, RCNNs, and Vision Transformers.

- Hands-on experience with deep learning frameworks : PyTorch / TensorFlow, and Keras.

- Practical experience in optimizing models for deployment using ONNX, TFLite, and CoreML.

- Experience in model compression techniques : quantization, pruning, and knowledge distillation.

- Strong grasp of Python and libraries such as OpenCV, NumPy, Pandas, scikit-learn.

- Familiarity with GPU computing and libraries like CUDA, cuDNN, and TensorRT.

- Exposure to speech signal processing and audio embedding is a plus.

- Knowledge of JavaScript is an added advantage.

- Self-driven, adaptable, and capable of working independently and in a collaborative environment.