S
Sesame Tech Stack
AI-powered wearable hardware with custom embedded OS and inference stack
Sesame is building consumer wearables with on-device AI, running a custom RTOS + embedded OS (Zephyr, FreeRTOS) paired with inference acceleration (vLLM, SGLang, PyTorch, JAX, TensorFlow). The stack reveals a company solving real-time ML inference at the edge: they're adopting NetSuite and SAP to manage hardware supply chains while grappling with model initialization latency, audio pipeline optimization, and sensor integration at scale. Engineering dominates the org (38 roles open), with active hiring in Taiwan signaling manufacturing readiness for volume production.
Tech Stack 59 technologies
Core StackPython PyTorch AWS Kubernetes TensorFlow RTOS C/C++ AOSP Zephyr FreeRTOS DSP Bluetooth USB Qualcomm QNX Assembly JTAG GDB vLLM SGLang GCP Azure Ray iOS iOS SDK WebRTC WebSockets Bluetooth Low Energy Jax FPGA+29 more
What Sesame Is Building
◆Challenges
- Reducing model initialization times
- Scaling inference workloads
- Real-time audio pipelines
- Integrating complex sensing systems into wearables
- Supplier risk mitigation
- Scalability challenges across domains
- Integration challenges
- Tight timelines
- Optimizing test coverage
- Optimizing sensor performance for mass production
▲Active Projects
- Wearable consumer product development
- Custom embedded os for ai-driven real-time consumer applications
- Supply risk mitigation
- Turbocharge serving layer
- Npi supplier readiness
- Manufacturing test strategy for new product introductions
- Test fixture and equipment design
- Firmware development workflow improvement
- End-to-end hardware development process
- Audio capture and render systems
Hiring Activity
70 roles · 25 in 30d
Department
Seniority
Company intelligence
