About Embedl

Embedl is a fast-growing technology start-up based in Gothenburg, Sweden. With a team of over 20 dedicated professionals, we are on a mission to make AI affordable and efficient in embedded systems. Our development team is headquartered in Gothenburg, and our innovative solutions serve customers all over the world. We maintain close connections with academic research institutions, industry leaders, and the innovation ecosystem, supported by strong relationships with venture capital partners.

The Role

As Embedded AI Specialist, you own the last mile between state-of-the-art models and the silicon that runs them. Your mission is to integrate, optimize, and profile the capabilities of Embedl’s SDK on a widening spectrum of edge accelerators.

Responsibilities

• Build & maintain inference backends for the Embedl SDK's growing set of supported edge accelerators including NVIDIA TensorRT, Qualcomm QNN, Ambarella ADK, Texas Instruments TIDL, etc.
• Continuously track the state of the art in embedded-AI workflows. Distill the best compiler flags, memory layouts, and optimization tricks, and share them across our engineering and research teams.
• Profile, benchmark, and resolve bottleneck in models. Pinpoint performance hot-spots, then deliver targeted optimizations that turn good benchmarks into great real-world latency and power figures.
• Extend Embedl’s model-optimization capabilities. Sparsity, Quantization, and more works on an ever-broader range of neural-network architectures and hardware platforms.

Who You Are

An engineer comfortable reasoning at every layer from high-level neural-network graphs down to low-level instructions. You thrive on unraveling the hardware details that turn good models into great products.

Qualifications

Must-Have

• B.Sc./M.Sc. in Computer Engineering, Electrical Engineering, Computer Science, or equivalent hands-on experience.
• Demonstrated success porting or optimizing deep-learning inference on at least one accelerator family (e.g., Qualcomm, TI, NVIDIA Jetson, ARM Ethos-U/NPU).
• 3+ years of experience in C/C++ and Python for tooling and automation.
• Solid understanding of compiler toolchains, quantization, and memory/performance trade-offs.

Nice-to-Have

• Experience writing custom CUDA or Triton kernels.

What We Offer

Collaborative Culture: Join a high-performance team that values teamwork, knowledge sharing, and a positive working environment.

Professional Growth: Opportunities for continuous learning, mentorship, and career advancement.

Benefits: Health and wellness allowance (friskvårdsbidrag) of 5000 SEK and private health insurance.

Application Process

The selection and interview process is ongoing, so please send your application in English as soon as possible. For any questions or clarifications, reach out to Ann Tornberg at ann@embedl.com.

You are welcome at Embedl for who you are, no matter where you come from, what you look like, or what your favorite IDE is.

Join us at Embedl and be a part of shaping the future of efficient AI!