Scientific Software Engineer
Fathom is working to create hardware for the future of machine intelligence through the development of optical hardware that computes with light. Our optical processors will enable training of human-brain-scale neural networks with unprecedented performance. We’re a venture backed startup, based in Palo Alto, CA.
We’re seeking talented software engineers with strong interests in machine learning and an understanding of neural networks or background in programming in some area of science.
What you’d do:
- Give programming support to the optics team including developing sophisticated ray and physical optics models for new prototype systems.
- Hardware and software instrumentation for control and evaluation of optical computing systems
- Benchmark different designs
- Help develop, adapt and map general machine learning algorithms based on features of our hardware
- Work with our ML team to help implement cutting-edge machine learning algorithms on simulations of next generation systems, optimizing for time and energy.
- Excellent programming skills and experience in building APIs (Python, C/C++ and whatever becomes necessary)
- Deep passion and fundamental understanding of design, algorithms, and data structures
- Understanding of the fundamentals of neural networks (like implementing various neural networks by hand).
- Familiarity with computer architecture and implementing algorithms on CPUs, GPUs
- Some knowledge of current frameworks such as TensorFlow, Theano, etc.
- High creativity and productivity
- Efficient and effective written and verbal communication