Neural AI
R&D Embedded Software Engineer
1
Which company are you aiming at?
To generate precise rejection-prevention feedback and application strategy, tell us where you're targeting.
Apple
NVIDIA
Tesla
Qualcomm
Bosch
Garmin
2
Send your application
You need an account so we know where to send the feedback.
The job description
Tech stack. C/C++, Python, rapid prototyping, novel architectures (RISC-V, chiplets), research papers to prototype, FPGA prototyping, performance modeling
About the role
You will join the advanced technology group of a semiconductor company, the team that prototypes what the product roadmap will look like three to five years out. You will turn research ideas into working embedded prototypes: new compute architectures, novel memory hierarchies, and experimental software stacks. This role matters because the company's future products are decided by what this team proves possible, and a prototype that runs is worth a hundred slide decks.
What you will achieve
- Ship three working research prototypes per year, each demonstrating a novel embedded capability on FPGA or early silicon with measured results, not simulations alone, and demoed to product leadership each quarter
- Drive a 2x performance-per-watt improvement hypothesis from paper to measured prototype on a RISC-V vector extension experiment, with full methodology documented, and results peer-reviewed within the group
- Reduce prototype iteration time from months to weeks by building a reusable bring-up and benchmarking harness for experimental targets that the whole group adopts, measured from approval of the idea to first measurements
- Build the technical case (measured data, not projections) that moves at least one research project into the product roadmap within two years
- Publish or patent the genuinely novel work: one publication or filed patent per year as a team output you contribute to, with you as lead author or inventor at least once
What you will bring
Must-haves
- 4-7 years in embedded software with a track record of prototyping on unfamiliar or immature hardware
- Strong C/C++ plus Python for rapid experimentation, performance modeling, and data analysis
- Experience reading research papers and converting algorithms into working implementations on real targets
- Comfort with incomplete documentation, buggy toolchains, and hardware that barely works: you make progress anyway
- Knowledge of computer architecture fundamentals: pipelines, caches, memory hierarchies, and interconnects
- Experience with FPGA prototyping flows or cycle-accurate simulators for pre-silicon software work
- Intellectual honesty: you kill your own ideas when the measurements say no, and you document why
Nice-to-haves
- RISC-V ecosystem experience including custom extensions or toolchain work
- Background in compilers or runtime systems for novel architectures
- Prior publications, patents, or open-source research prototypes with real users