Deep Learning Engineer

Role Overview

We are seeking a skilled Deep Learning Engineer to design, train, and deploy advanced neural network architectures. This role goes beyond fine-tuning pretrained models, requiring hands-on experience building deep learning models from scratch. You will work on high-impact problems across areas such as computer vision, NLP, and multimodal AI, contributing directly to the development of innovative AI products.

Responsibilities

• Design, implement, and train deep neural networks (e.g., CNNs, RNNs, Transformers, Diffusion Models) from scratch.
• Select and optimize loss functions, training algorithms, regularization techniques, and data augmentation strategies.
• Analyze model performance using metrics, visualizations, ablations, and interpretability tools.
• Write clean, modular, and efficient code for research experiments and production deployment.
• Collaborate with ML researchers, data engineers, and product teams to bring models into production.
• Stay updated with the latest AI/ML research and assess innovations for adoption.
• Conduct internal benchmarks and document model development processes.

Requirements

• 3+ years of experience in training deep learning models end-to-end, beyond pre-trained fine-tuning.
• Strong programming skills in Python, with hands-on experience in PyTorch or TensorFlow.
• Solid understanding of neural network theory: backpropagation, activation functions, initialization, and optimization.
• Experience training on large datasets, including distributed training and GPU acceleration.
• Ability to design data pipelines and manage the model lifecycle in production environments.
• Strong intuition for debugging training instabilities, overfitting, or mode collapse.
• Bachelor's Degree.

Nice to Have

• Productionizing models using tools like ONNX, TensorRT, or quantization.
• Knowledge of LLMs, diffusion models, or custom neural architectures.
• Contributions to open-source ML projects or published research papers.
• Experience working with multimodal inputs (e.g., vision + text, audio + text).

Skills

• Python
• PyTorch
• TensorFlow
• CNNs
• Transformers