- Bring up, test and debug neural networks using ML frameworks like TensorFlow, PyTorch, ONNX etc. for Blaize-GSP.
- Bring up and enhance TVM features
- Analyze and enhance efficiency & stability of neural networks.
- Develop & maintain Model Conversion Tool software stack for Blaize-GSP.
- Network Optimization, Node fusion, graph lowering, adding custom operations, profiling & performance fine tuning.
- Model Quantization
- MS or BS in computer science or Electronics
- 8-12yrs years of relevant experience or an equivalent combination of experience and education.
REQUIRED KNOWLEDGE, SKILLS, AND ABILITIES
- Hands on experience in C,C++,Python, NumPy, ML frameworks like TensorFlow, PyTorch, ONNX and others.
- Good knowledge of Linear algebra
- Knowledge of NWs optimization, graph lowering and finetuning
- Good analytical skills
- Good understanding of algorithms, OOPS concepts and SW Design Patterns.
- Good debugging skills.
- Strong knowledge of TVM FW
- Experience or knowledge in HW Architecture is an added advantage.
- Experience on full stack framework development like any of Multimedia frameworks, GStreamer, OpenVx, OpenMax, OpenGL, OpenGL-ES, Vulkun, Mesa, etc. is a plus.
- Experience on driver development on Linux platform.
- CLC/assembly compute kernels for the Blaize-GSP.
MANDATORY SKILLS
- Hands on experience in C/CPP, Python, NumPy,openCLC++
- Hand on experience in ML frameworks like TensorFlow, PyTorch and ONNX
- Hand on experience in TVM
