ML Hardware Verification Manager
- Discuss this with your agent
- Open in Claude
- Open in ChatGPT
Prepare application
- Draft this with your agent
- Open in Claude
- Open in ChatGPT
Role details
Tech stack
+5 more
Job description
Meta is building custom silicon to power the next generation of AI and machine learning workloads across its Reality Labs products. We are seeking an experienced ML Hardware Verification Manager to lead the design verification efforts for custom ML accelerators and silicon platforms. In this role, you will manage a team of hardware verification engineers responsible for ensuring the functional correctness, coverage closure, and silicon readiness of ML-specific hardware designs, including tensor processing units, memory subsystems, and interconnect fabrics. You will partner closely with ML researchers, architects, RTL designers, and software teams to define verification strategies and drive tape-out milestones., 1. Lead and manage a team of hardware design verification engineers working on custom ML accelerator silicon, setting technical direction and driving execution against tape-out schedules
- Define and own the verification strategy for ML hardware blocks, including compute engines, memory hierarchies, and on-chip interconnects, encompassing simulation and formal verification
- Drive functional coverage closure and sign-off criteria across multiple silicon projects, ensuring verification completeness before tape-out
- Partner with RTL design and architecture teams to review specifications, identify verification risks early, and influence design-for-verifiability decisions
- Collaborate with post-silicon validation and software teams to align pre-silicon verification environments with bring-up and debug infrastructure
- Establish and scale verification infrastructure, including UVM testbench development, constrained-random stimulus generation, and assertion-based verification frameworks
- Translate organizational strategy into team roadmaps, balancing short-term tape-out deliverables with long-term investments in verification automation and reuse
- Support SW-HW co-design, development of software unit tests and/or custom kernels
- Communicate verification status, risk assessments, and mitigation plans clearly to engineering leadership and cross-functional stakeholders
Requirements
- 12+ years of experience in hardware design verification, including functional simulation, coverage-driven verification, and formal verification of digital logic
- 5+ years of experience in managing hardware engineering teams, including setting technical direction and driving execution across multiple concurrent projects
- Experience in Machine Learning on IPs and Silicon development
- Knowledge of SystemVerilog Assertions, X-propagation, UPF-power aware, gate-level simulations
- Experience with UVM-based testbench architecture and constrained-random verification methodologies applied to complex SoC or accelerator designs
- Experience in driving verification closure for tape-out on custom silicon, including coverage analysis, regression management, and sign-off processes
- Experience collaborating with RTL design, physical design, and post-silicon validation teams across the full silicon development lifecycle
- BS degree in Electrical Engineering, Computer Science, or relevant field, 18. Knowledge of convolutional neural networks, transformers, LLMs
- Master’s or PhD degree in Electrical Engineering, Computer Science, or relevant field
- Familiarity with emulation platforms and FPGA-based prototyping as part of a pre-silicon verification flow
- Experience with SW/HW co-design, firmware/HW interaction testing
- Experience in verifying ML accelerator hardware components such as matrix multiply units, dataflow engines, or high-bandwidth memory interfaces
Apply for this position
This job is hosted externally. Click below to view the full posting and apply.
Prepare application
- Draft this with your agent
- Open in Claude
- Open in ChatGPT
Good distractions
Talks and stories from around this role — technically off-topic, practically not.
Moments
Explore playlistsVideos
See allRelated articles
See all
What Are Large Language Models?
How to Become an AI Engineer
MLOps And AI Driven Development
How We Built a Worry-Free System That Runs for 10+ Years – And What We’d Do Again