AI Hiring Index

SambaNova · Infrastructure · Staff+ · Posted 2026-08-17

Principal Design Verification Engineer

SambaNova · San Jose, California, United States · $180k–255k base

This range's midpoint is above 49% of posted infrastructure ranges at AI companies right now. See the salary index.

Apply on SambaNova's site Watch SambaNova for new roles

SambaNova is a leader in next-generation AI infrastructure, delivering a full-stack inference platform for customers worldwide. At the core of SambaNova's technology is the RDU (Reconfigurable Dataflow Unit) — a chip built on a dataflow architecture rather than the traditional GPU model. Its decode performance is especially strong for agentic workloads like multi-turn agents, code generation, and long-running applications. RDUs are packaged into SambaRack, rack-scale hardware that lets customers deploy state-of-the-art models with better performance, greater energy efficiency, and faster time to value.

About the team

SambaNova's RDU is the processor behind our inference systems, and design verification ensures quality and reliability before production. The team owns verification for block, full-chip, and gate-level benches, performance verification, coverage closure, and the emulation and FPGA prototypes.

About the role

SambaNova is looking for a Principal Design Verification Engineer to own verification for  our next-generation RDU, the reconfigurable dataflow silicon behind our machine learning and AI systems. This is a senior individual-contributor role where you'll own the test plan, testbench architecture, coverage model, and sign-off criteria, while shaping the team-wide infrastructure, reuse patterns, and verification practices.

You'll work alongside principal engineers from the design, architecture, compiler, and runtime teams who have built and evolved this verification environment across several generations of silicon, to verify the chip against production ML workloads.

Responsibilities

Some of your responsibilities will include...

Own verification end to end for a major subsystem of the RDU: test plan, testbench architecture, coverage model, and sign-off criteria

Build the coverage model for your area, set the closure bar, and drive it to closure ahead of tape-out

Develop testbench components, agents, checkers, and stimulus at block, full-chip, system, and gate level

Serve as the technical authority for verification of your subsystem: review and arbitrate testbench design decisions, and lead debug on the hardest failures

Own regression health and failure triage for your area: drive root cause, not just reproduction

Verify functional and performance behavior

Advance DV methodology beyond your own area: reusable infrastructure, verification IP integration, and AI-assisted verification workflows the rest of the team adopts

Contribute to emulation and FPGA prototyping, and support silicon bring-up and post-silicon debug in the lab

Set the technical bar through design and DV reviews, and mentor engineers to raise the level of the team's verification practice

Required qualifications

BS or MS in EE, CS, or CE with 7+ years (or equivalent) of industry experience in ASIC design verification

Deep hands-on expertise in SystemVerilog and SVA with a standard methodology (UVM, OVM, or VMM), plus scripting in Python and C/C++

Full-cycle DV ownership from test plan through tape-out on at least one complex ASIC or major subsystem, including owning a coverage model to closure

Experience verifying digital logic in a CPU, GPU, SoC, or comparably complex ASIC

Track record of debugging hard, intermittent, or low-reproducibility pre-silicon failures to root cause

Demonstrated technical leadership without formal authority: setting direction in reviews, arbitrating testbench decisions, and mentoring engineers

Working knowledge of modern computer architecture and of machine learning accelerators or SoCs

Preferred qualifications

MS or PhD in EE, CS, or CE with 10+ years of experience in ASIC design verification

Depth in high-speed I/O verification: protocol-level verification of PCIe or Ethernet, and memory subsystems such as HBM or DDR

Performance verification methodology: building performance benches and analyzing results, not only functional correctness

Formal verification and …

New infrastructure roles at AI companies, every Monday. The week's openings in this function across 286 companies, plus the weekly index. Free.

More infrastructure roles at SambaNova

See also: Hardware Engineer jobs · AI jobs in San Francisco Bay Area · SambaNova salaries · Python jobs · C++ jobs · Agents jobs.

This listing is reproduced from SambaNova's public careers feed and links to the original. AI Hiring Index is not the employer and does not accept applications. All SambaNova roles · AI salaries.