Tensordyne
Designs AI chips and systems for efficient model inference.
Company facts
- Founded
- 2017
- Employees
- 117
- Headquarters
- United States
- Type
- Private
- Funding
- Series C · $35M · 2025
Open positions
Senior Software Engineer, Inference Performance
About Tensordyne AI is transforming our world. It can perform cognitive functions that previously only humans could do, such as perceiving interactions across different modalities and environments - with the ability to…
Working Student - LLM Inference Performance
About Tensordyne AI is transforming our world. It can perform cognitive functions that previously only humans could do, such as perceiving interactions across different modalities and environments - with the ability to…
Senior Software Engineer — ML Inference & Numerics
About Tensordyne AI is transforming our world. It can perform cognitive functions that previously only humans could do, such as perceiving interactions across different modalities and environments - with the ability to…
System Software/Embedded Engineer (Munich)
Embedded/systems engineer on Tensordyne's Platform Software team, building firmware, bootloaders/BSPs, device drivers, and Linux kernel modules for the company's generative-AI inference hardware, including board bring-up, debugging/profiling tooling, and performance/power optimization alongside silicon and hardware teams.
Working Student - LLM Quantization & Inference Performance
A working student in Munich who researches how low-precision quantization affects LLM quality and inference speed: running evaluations, implementing post-training quantization methods in Python/PyTorch, and building reproducible tooling alongside quantization, compiler, and runtime engineers optimizing models for Tensordyne hardware.
Software Engineer – Cloud Services – Rust
Builds distributed cloud services in Rust for Tensordyne's generative AI inference platform — developing the inference backend software stack, exposing APIs for generative AI, maintaining in-house Rust code, and orchestrating agent teams. Based in Munich; modern Rust, async patterns, HTTP, and Kubernetes experience preferred.
Ops/Developer Platform Engineer
Member of Tensordyne's DevOps/MLOps team building and maintaining the internal developer platform: CI pipelines for the monorepo, ML workflow orchestration, cloud dev/build environments, and Rust-based internal developer services, all running on GCP with Kubernetes, Terraform, and agentic workflows.