Engineering Manager, AI Compiler Analysis

at Nvidia

USD 168,000-322,000 per year

SENIOR

✅ On-site

Tech Stack
Tag name is followed by "@" symbol and proficiency level value. About proficiency levels:

1-2 — basic awareness. Minimal hands-on experience, and a rudimentary understanding of the technology's purpose;

3-6 — daily use. Comfortable and regular usage, capable of handling common tasks and challenges related to the technology;

7-9 — you are an expert, you can teach others, you know all the pitfalls and tricks;

10 — exceptional knowledge, comprehensive understanding, and adeptness in all aspects of the technology, including advanced problem-solving. Think twice before claiming or demanding such level.

AI @ 4 Agentic AI CUDA Debugging @ 4 Deep Learning @ 4 GPU Hiring @ 7 JAX @ 4 LLM Machine Learning @ 6 Mentoring @ 7 People Management @ 7 PyTorch @ 4 TensorRT @ 4

Details

NVIDIA's compiler technologies enable AI at scale by transforming rapidly evolving deep learning models into optimized GPU programs for training and inference. This hands-on compiler engineering leadership role will lead the strategy for verifying AI compilers used by next-generation deep learning workloads, addressing challenges across compilers, machine learning frameworks, numerical computing, formal reasoning, and large-scale systems engineering.

Responsibilities

Lead, mentor, and grow a highly technical team responsible for AI compiler verification.
Own verification of next-generation AI workloads, including large language models and agentic AI systems, across the AI compiler stack and execution pipeline.
Define formal-verification requirements for AI compiler transformations and generated GPU programs, including formal specifications, tensor and operator semantics, semantic preservation, code equivalence, numerical behavior, and properties stressed by AI-generated or adversarial workloads.
Drive AI-assisted and compiler-aware verification techniques, including adversarial workload generation, differential testing, symbolic reasoning, formal methods, fuzzing, static analysis, and automated debugging.
Partner with AI compiler development, CUDA software, machine learning framework, runtime, product, and AI software teams to build scalable verification infrastructure, improve engineering velocity, and increase production confidence.

Requirements

Bachelor's, master's, or doctoral degree in Computer Science, Computer Engineering, or a related field, or equivalent experience.
At least 10 years of relevant software engineering experience, including at least 3 years leading engineering teams or major technical initiatives.
Experience with AI compiler or framework technologies such as MLIR, TensorRT, XLA, Triton, PyTorch, or JAX.
Fluency with AI workload and machine learning framework concepts, including computation graphs, tensor operations, model execution, and training or inference workflows.
Strong people management skills, including hiring, mentoring, performance management, and team development.

Preferred Qualifications

Hands-on experience with deep learning compiler internals, including compiler intermediate representations, optimization and lowering pipelines, code generation, runtime integration, or production compiler infrastructure.
Experience verifying performance-sensitive compiler behavior and investigating subtle regressions in production AI and machine learning systems using computational methods, fuzzing, code inspection, or automated debugging.
Background in formal verification or programming languages, with familiarity with formal specifications, theorem proving, Lean, SMT/SAT solvers, or symbolic reasoning.

Compensation and Benefits

Base salary range of $168,000–$270,250 for Level 2.
Base salary range of $200,000–$322,000 for Level 3.
Eligibility for equity and benefits.
Salary is determined based on location, experience, and compensation for similar positions.
Applications will be accepted at least until August 3, 2026.
NVIDIA is an equal opportunity employer and is committed to an inclusive work environment.