Software Engineer, LLVM Compiler - MTIA

Meta; Menlo Park (CA); On-site; USD 190,000 - 260,000; Full time

Job summary

Meta is hiring a Software Engineer to own GenAI inference on MTIA, spanning serving frameworks, distributed inference, graph-mode execution, and PyTorch integration.

You will set technical direction, drive multi-quarter programs across teams, and own quality in accuracy, stability, and test coverage on custom silicon.

Qualifications

• Bachelor's degree in CS, CE, or related field.
• 8+ years in systems software, ML infra, or compiler/framework development.
• 5+ years in generative AI inference or production model-serving.
• Proficient in Python, C++, or Rust with low-level code.
• Experience with GenAI inference optimization and latency/throughput tradeoffs.
• Experience with runtime graph-mode execution and memory management.
• Led design and end-to-end delivery of frameworks or infra projects.
• Experience across framework, compiler, runtime, and hardware layers.
• Familiarity with PyTorch compilation stack and custom operator integration.
• Strong debugging for numerical accuracy and stability.

Responsibilities

• Own GenAI inference area spanning serving frameworks and distribution.
• Set technical direction and ensure accuracy, stability, performance, and test coverage on custom silicon.

Skills

Python/C++/Rust

GenAI optimization

DL framework internals

Runtime optimization

Technical design leadership

Cross-layer system knowledge

Numerical accuracy debugging

LLM inference serving

Responsible AI practices

Long-context inference

AI accelerator platforms

Production performance tuning

Tool integration for AI workflows

Distributed inference

Quantization

Ongoing AI learning

PyTorch compilation stack

Education

Bachelor's degree in CS/CE

Equivalent practical experience

Job description

Meta designs and deploys its own AI systems. MTIA, the Meta Training and Inference Accelerator, is Meta's family of in-house AI accelerator ASICs, deployed in production across Meta's data centers and expanding workload coverage from ranking and recommendation to generative AI. The MTIA Software team is part of the AI & Compute Foundation organization. Because the hardware is ours, the software is ours too: we build the entire stack a chip vendor would normally supply: compiler and toolchain, runtime, kernel libraries, developer tooling, and deep PyTorch integration, co-designed with the silicon teams generation over generation.

The GenAI Frameworks org owns the layer where a model meets the machine. We take frontier models, including large language, multimodal and diffusion models, and make them run efficiently on MTIA. Our work determines how fast those models respond, how much context they can afford to hold, how many of them a given amount of silicon can serve, and how quickly a newly released model can be running and trusted in production.

We are hiring a Software Engineer to own a key area of GenAI inference on MTIA, spanning serving frameworks, distributed inference, graph-mode execution and compilation, and core PyTorch integration. This is a domain-expert role: you will set technical direction for your area, drive ambiguous multi-quarter programs across team boundaries, and own the quality bar for accuracy, stability, performance, and test coverage on custom silicon.

• Bachelor's degree in Computer Science, Computer Engineering, relevant technical field, or equivalent practical experience Bachelor's degree in Computer Science, Computer Engineering, a relevant technical field, or equivalent practical experience
• 8+ years of professional experience in systems software, ML infrastructure, performance engineering, or compiler/framework development
• 5+ years of hands-on experience with generative AI inference or training frameworks, or production model-serving systems
• Proficiency in Python, C++, or Rust, including low-level systems code and performance-critical paths
• Experience with GenAI inference optimization: prefill and decode optimization, KV-cache management and compression, batching and scheduling, and reasoning about latency and throughput tradeoffs
• Hands-on experience with deep learning framework internals beyond model authoring — operator registration and dispatch, eager execution, autograd, or graph capture and compilation paths
• Experience with runtime-level optimization: graph-mode execution, host-side latency reduction, memory allocation and placement
• Demonstrated experience leading technical design and end-to-end delivery of frameworks or infrastructure projects, including test strategies and CI, across team boundaries
• Experience working across multiple layers of a system stack (framework, compiler, runtime, hardware) and root-causing issues across component boundaries
• Experience debugging numerical accuracy issues and defending accuracy and stability bars, not just performance
• Experience with LLM inference serving stacks, including continuous batching, paged or chunked attention, speculative decoding, prefix caching, disaggregated prefill and decode, or mixture-of-experts routing and load balancing
• Experience adhering to and implementing responsible, ethical AI practices (e.g., risk assessment, bias mitigation, quality and accuracy reviews)
• Experience with long-context inference techniques: context and sequence parallelism, ring or streaming attention, hierarchical or offloaded KV cache
• Experience with AI accelerator platforms (GPU, TPU, or custom ASICs) and bringing up a new hardware backend in a major machine learning framework
• Experience with performance optimization in production environments, with a track record of using profiling and tracing to understand and resolve performance bottlenecks
• Demonstrated ability to integrate AI tools to optimize/redesign workflows and drive measurable impact (e.g., efficiency gains, quality improvements)
• Experience with distributed inference or training at scale: tensor, pipeline, and expert parallelism, collective communication primitives, RDMA transports, and multi-host serving, scale-up and scale-out design
• Experience with low-precision numerics and quantization (FP8, block-scaled formats, INT8/INT4), including calibration and accuracy error analysis, debugging tools for distributed systems
• Demonstrated ongoing AI skill development (e.g., prompt/context engineering, agent orchestration) and staying current with emerging AI technologies
• Familiarity with the PyTorch compilation stack (TorchDynamo, FX IR, Torch Inductor, dynamic shapes, Triton, AOT I) and custom operator integration