Snap Inc is a technology company. We believe the camera presents the greatest opportunity to improve the way people live and communicate. Snap contributes to human progress by empowering people to express themselves live in the moment learn about the world and have fun together. The Companys three core products are Snapchat a visual messaging app that enhances your relationships with friends family and the world; Lens Studio an augmented reality platform that powers AR across Snapchat and other services; and its AR glasses Spectacles.
Were looking for an Neural Processing Core Architect to join our team at Snap Inc!
You will play a key role in defining and developing next-generation processor architectures for real-time Edge AI enabling intelligent and adaptive capabilities on AR Glasses. Working from our Eindhoven office youll collaborate closely with ML researchers hardware engineers and system architects worldwide to build processors that are deeply co-optimized for modern neural workloads from super-resolution to compact language models/state-space models (LLMs NLMs SLMs and state-space/Mamba). In addition your role will involve:
Architect and prototype processor cores and compute subsystems specialized for AI workloads in ultra-low-power real-time environments.
Co-design neural network models and training methodologies that exploit hardware efficiency including quantization-aware training sparsity and mixed-precision approaches.
Define and carry out performance analysis on dataflows instruction sets and memory architectures optimized for end-to-end learning and inference on-device.
Collaborate with machine learning teams to identify key model characteristics and map them to architectural primitives.
Conduct performance bandwidth and power modeling to guide design decisions for AI processors and SoC integration.
Prototype benchmark and iterate on architectures for super-resolution generative and/or language-based AI running at the edge.
Work with cross-functional teams to integrate processor designs into hardware platforms and software toolchains.
Contribute to Snaps Edge AI roadmap helping define the future of compute for always-on intelligent systems.
Knowledge Skills & Abilities:
Deep understanding of computer architecture principles especially as applied to AI accelerators and embedded processors.
Strong knowledge of neural networks including training optimization for hardware constraints (e.g. quantization pruning).
Experience with ML frameworks (PyTorch and/or TensorFlow).
Familiarity with hardware modeling tools (eg SystemC) and performance profiling.
Solid grasp of AI workloads such as super-resolution diffusion language and/or multimodal perception models.
Excellent communication and collaboration skills.
MSc or PhD in Electrical Engineering Computer Engineering or a related field (PhD research may count as professional experience).
Proven experience in processor or accelerator architecture design
Experience with AI model optimization.
Experience with hardware design (e.g. Verilog) and optimization.
Strong analytical programming and documentation skills.
Experience designing hardware-aware training and/or neural architectures optimized for embedded compute targets.
Familiarity with hardware/ML co-design.
Track record of research and/or publications in hardware-aware AI design and/or hardware design for AI.
If you have a disability or special need that requires accommodation please dont be shy and provide us some information.
Default Together Policy at Snap: At Snap Inc. we believe that being together in person helps us build our culture faster reinforce our values and serve our community customers and partners better through dynamic collaboration. To reflect this we practice a default together approach and expect our team members to work in an office 4 days per week.
At Snap we believe that having a team of diverse backgrounds and voices working together will enable us to create innovative products that improve the way people live and communicate. Snap is proud to be an equal opportunity employer and committed to providing employment opportunities regardless of race religious creed color national origin ancestry physical disability mental disability medical condition genetic information marital status sex gender gender identity gender expression pregnancy childbirth and breastfeeding age sexual orientation military or veteran status or any other protected classification in accordance with applicable federal state and local laws. EOE including disability/vets.
We believe the camera presents the greatest opportunity to improve the way people live and communicate.