M
Software Engineer, MTIA SW Performance Autotuning
Accepting applicationsMeta · Menlo Park, CA
Full-Time Mid_senior AIASICC++Pythonmachine learning
Posted
1d ago
Category
Design
Experience
Mid_senior
Country
United States
We are looking for an experienced engineer to lead performance autotuning on MTIA — Meta's custom training and inference accelerator. You will lead the MTIA Software Performance Autotuning team (part of Infra Foundations) and own how we extract maximum performance from our hardware, automatically and at scale.Every kernel, every compiled graph, and every runtime configuration has a large space of possible implementations — tile sizes, scheduling, memory layouts, fusion decisions, precision choices — and the right one depends on the chip, the model, and the shape. Hand-tuning does not scale. The team's core mission is to make MTIA fast by default: building the search infrastructure, cost models, and tuning methodology that finds the best configuration without a human in the loop. As a technical leader, you will define our autotuning strategy, architect the search and benchmarking infrastructure, and partner closely with compiler, kernel, runtime, and product (e.g., Ads Ranking, Recommendation Systems, GenAI) teams to turn performance headroom into shipped speedups.Autotuning operates across the full MTIA software stack — FX graphs, compiler, kernels, runtime, PyTorch — which means lots of cross-team collaboration. We partner closely with machine learning engineers across Ads, Instagram/Facebook, and Meta Superintelligence Labs teams whose models run on MTIA.
Software Engineer, MTIA SW Performance Autotuning Responsibilities:
Own the performance autotuning strategy for MTIA inference and training — current and future chip generations
Build and maintain our autotuning infrastructure: search spaces, cost models, benchmarking harnesses, and performance regression detection
Define how we trade off search cost against performance gains, and where autotuning fits in the compile/deploy pipeline
Partner with kernel and compiler teams to expose tunable parameters and encode them into an automated search
Debug performance issues that span multiple layers — kernel, runtime, compiler, or serving infra — and drive them to root-cause
Collaborate with product teams to accelerate model performance on MTIA
Set technical direction and help the team grow. We've scaled fast and need people who can elevate the impact around them, not just ship their own code
Minimum Qualifications:
Bachelor's degree in Computer Science, Computer Engineering, relevant technical field, or equivalent practical experience
Strong Python and C++ skills, with hands-on experience across the PyTorch stack
4+ years in ML systems, AI infra, performance engineering, or similar
Experience driving problems that span multiple teams, where no one owns the whole picture
Experience with accelerator performance concepts — roofline analysis, memory bandwidth, occupancy, and what makes kernels fast or slow on hardware
Track record of setting technical direction and mentoring engineers
BS in CS, CE, Math, or equivalent experience
Preferred Qualifications:
Experience adhering to and implementing responsible, ethical AI practices (e.g., risk assessment, bias mitigation, quality and accuracy reviews)
Demonstrated ability to integrate AI tools to optimize/redesign workflows and drive measurable impact (e.g., efficiency gains, quality improvements)
Demonstrated ongoing AI skill development (e.g., prompt/context engineering, agent orchestration) and staying current with emerging AI technologies
Experience with production ML models (recommender systems, LLMs, ranking) is useful but not required
Hands-on experience with torch.compile, TorchInductor, or other ML compiler stacks (XLA, TVM, MLIR, Triton)
Experience with autotuning, cost models, or search-based optimization (e.g., Ansor, AutoTVM, learned schedulers)
MS or PhD in CS, CE, compilers/systems, or related
Experience with hardware bring-up or accelerator development (GPU, TPU, or custom ASIC)
Experience with kernel-level performance optimization — tiling, scheduling, memory layout, fusion, and how they interact on real hardware
About Meta:
Meta builds technologies that help people connect, find communities, and grow businesses. When Facebook launched in 2004, it changed the way people connect. Apps like Messenger, Instagram and WhatsApp further empowered billions around the world. Now, Meta is moving beyond 2D screens toward immersive experiences like augmented and virtual reality to help build the next evolution in social technology. People who choose to build their careers by building with us at Meta help shape a future that will take us beyond what digital connection makes possible today—beyond the constraints of screens, the limits of distance, and even the rules of physics.
Meta is proud to be an Equal Employment Opportunity and Affirmative Action employer. We do not discriminate based upon race, religion, color, national origin, sex (including pregnancy, childbirth, or related medical conditions), sexual orientation, gender, gender identity, gender expression, transgender status, sexual stereotypes, age, status as a protected veteran, status as an individual with a disability, or other applicable legally protected characteristics. We also consider qualified applicants with criminal histories, consistent with applicable federal, state and local law. Meta participates in the E-Verify program in certain locations, as required by law. Please note that Meta may leverage artificial intelligence and machine learning technologies in connection with applications for employment.
Meta is committed to providing reasonable accommodations for candidates with disabilities in our recruiting process. If you need any assistance or accommodations due to a disability, please let us know at [email protected].
$154,003/year to $217,000/year + bonus + equity + benefits
Individual compensation is determined by skills, qualifications, experience, and location. Compensation details listed in this posting reflect the base hourly rate, monthly rate, or annual salary only, and do not include bonus, equity or sales incentives, if applicable. In addition to base compensation, Meta offers benefits. Learn more about benefits at Meta.
Show more Show less
Software Engineer, MTIA SW Performance Autotuning Responsibilities:
Own the performance autotuning strategy for MTIA inference and training — current and future chip generations
Build and maintain our autotuning infrastructure: search spaces, cost models, benchmarking harnesses, and performance regression detection
Define how we trade off search cost against performance gains, and where autotuning fits in the compile/deploy pipeline
Partner with kernel and compiler teams to expose tunable parameters and encode them into an automated search
Debug performance issues that span multiple layers — kernel, runtime, compiler, or serving infra — and drive them to root-cause
Collaborate with product teams to accelerate model performance on MTIA
Set technical direction and help the team grow. We've scaled fast and need people who can elevate the impact around them, not just ship their own code
Minimum Qualifications:
Bachelor's degree in Computer Science, Computer Engineering, relevant technical field, or equivalent practical experience
Strong Python and C++ skills, with hands-on experience across the PyTorch stack
4+ years in ML systems, AI infra, performance engineering, or similar
Experience driving problems that span multiple teams, where no one owns the whole picture
Experience with accelerator performance concepts — roofline analysis, memory bandwidth, occupancy, and what makes kernels fast or slow on hardware
Track record of setting technical direction and mentoring engineers
BS in CS, CE, Math, or equivalent experience
Preferred Qualifications:
Experience adhering to and implementing responsible, ethical AI practices (e.g., risk assessment, bias mitigation, quality and accuracy reviews)
Demonstrated ability to integrate AI tools to optimize/redesign workflows and drive measurable impact (e.g., efficiency gains, quality improvements)
Demonstrated ongoing AI skill development (e.g., prompt/context engineering, agent orchestration) and staying current with emerging AI technologies
Experience with production ML models (recommender systems, LLMs, ranking) is useful but not required
Hands-on experience with torch.compile, TorchInductor, or other ML compiler stacks (XLA, TVM, MLIR, Triton)
Experience with autotuning, cost models, or search-based optimization (e.g., Ansor, AutoTVM, learned schedulers)
MS or PhD in CS, CE, compilers/systems, or related
Experience with hardware bring-up or accelerator development (GPU, TPU, or custom ASIC)
Experience with kernel-level performance optimization — tiling, scheduling, memory layout, fusion, and how they interact on real hardware
About Meta:
Meta builds technologies that help people connect, find communities, and grow businesses. When Facebook launched in 2004, it changed the way people connect. Apps like Messenger, Instagram and WhatsApp further empowered billions around the world. Now, Meta is moving beyond 2D screens toward immersive experiences like augmented and virtual reality to help build the next evolution in social technology. People who choose to build their careers by building with us at Meta help shape a future that will take us beyond what digital connection makes possible today—beyond the constraints of screens, the limits of distance, and even the rules of physics.
Meta is proud to be an Equal Employment Opportunity and Affirmative Action employer. We do not discriminate based upon race, religion, color, national origin, sex (including pregnancy, childbirth, or related medical conditions), sexual orientation, gender, gender identity, gender expression, transgender status, sexual stereotypes, age, status as a protected veteran, status as an individual with a disability, or other applicable legally protected characteristics. We also consider qualified applicants with criminal histories, consistent with applicable federal, state and local law. Meta participates in the E-Verify program in certain locations, as required by law. Please note that Meta may leverage artificial intelligence and machine learning technologies in connection with applications for employment.
Meta is committed to providing reasonable accommodations for candidates with disabilities in our recruiting process. If you need any assistance or accommodations due to a disability, please let us know at [email protected].
$154,003/year to $217,000/year + bonus + equity + benefits
Individual compensation is determined by skills, qualifications, experience, and location. Compensation details listed in this posting reflect the base hourly rate, monthly rate, or annual salary only, and do not include bonus, equity or sales incentives, if applicable. In addition to base compensation, Meta offers benefits. Learn more about benefits at Meta.
Show more Show less