Ai Software Engineer

**Bifrost**:
Bifrost generates 3D worlds at scale and high fidelity synthetic datasets for AI development. This next-generation capability has use cases across defence, space, gaming, metaverse, mobility and robotics. We believe autonomous systems and computer vision have the opportunity to drastically improve the quality of life for billions of people.

However, one of the hardest problems in building AI for the physical world is that it is difficult for developers to collect the data they require from the real world. Our intelligent tools enable developers to generate and simulate vast 3D worlds to build cutting-edge AI systems to solve the world’s hardest problems. All without the need for 3D expertise.

These tools are used by the world’s most advanced organizations to land a spacecraft on Mars, catch pirates at sea, detect drug-carrying submarines and numerous critical use cases.

At Bifrost, you are backed by Sequoia Capital, Lux Capital, Wavemaker Partners, Cap Vista, Hustle Fund, Champion Hill Ventures and other prominent investors. Your teammates hail from nuTonomy, Motional, Medios, NVIDIA, GovTech, Lucasfilm, Industrial Light & Magic, Grab, TAIGER, Arup, The Parentinc and other game-changing companies.

**Why Bifrost?**:
Build captivating and photorealistic 3D worlds

Bring the bleeding edge of research in AI, GANs, 3D graphics into production

Learn about AI, defence, space, gaming, metaverse, mobility and robotics ‍

World-class team, investors, advisors and customers

Competitive salary and comprehensive insurance ‍
- ️ Flexible and hybrid environment with no red tape/bureaucracy

Trendy shophouse office in CBD near great restaurants and bars

Committed to helping you grow personally and professionally

Inclusive, judgement-free and a safe space to be yourself

Take ownership and build your skillset

Safe space to share dank memes

Fast growing company

**AI Software Engineer**:
**️‍ ️ Your Main Quest**:
We're looking for teammates excited to learn, build, and invent technologies that push the boundaries of computer vision and machine learning engineering.
- Develop, optimize, and scale libraries for synthetic data, deep learning model training and testing, and failure analysis
- Implement state-of-the-art computer vision and synthetic data research in production
- Implement and develop best practices for ML engineering
- Strategize and brainstorm new product features and direction

**Requirements**:

- Experience designing, writing, testing and documenting clean code, using version control, and proficiency translating business requirements into technical specifications
- Knowledge of deep learning computer vision architectures for classification, object detection, semantic segmentation, including convolutional neural networks, generative adversarial networks, auto-encoders, vision transformers, and others
- Able to implement, test, and debug models and pipelines in PyTorch or TensorFlow
- Able to analyze model weights, feature maps, and latent spaces to understand model sensitivity, performance, and failure cases
- Proficiency with the Python machine learning/vision stack — Pandas, Scipy, OpenCV, Scikit-learn, etc
- Strong foundation in the scientific method. Able to formulate hypotheses and carry out tests to verify them
- Able to communicate ideas and details clearly and effectively
- Experience - 1 to 4 years
- Location: Singapore or Remote

**✨ Bonus**:

- Experience with training frameworks such as PyTorch Lightning
- Experience with MLOps tools such as MLFlow, W&B, KubeFlow, etc
- Experience with multi-GPU and multi-node training and inference
- Experience with model optimization, pruning, and quantization
- Experience with deep learning acceleration frameworks such as JAX and CUDA
- Experience with autonomous vehicle perception, sensor fusion, and/or geospatial systems

**Sounds like you? **: