Research and Pathfinding Internship: AI Workload Compiler Optimization for CPU and GPU

Intel's Datacenter and AI Software Pathfinding team focuses on advancing compiler infrastructure for heterogeneous AI workloads, targeting both CPU and GPU architectures. The Software Team develops strategies for client and data center software, delivering specialized NPU and GPU IP, and supports CPU and GPU/accelerator roadmaps including integrated and discrete graphics.

Required Qualifications:

• Experience with compiler internals or programming languages (IR design, optimization passes)
• Programming skills in Python and C++ (MLIR/LLVM ecosystem desirable)
• Familiarity with CPU architecture including cache hierarchies and SIMD/vector instructions
• Current student of bachelor, master, or PhD studies in Computer Science, Electrical Engineering, or related field

Preferred Qualifications:

• Understanding of algebraic rewrite systems and/or e-graphs
• Prior work with LLVM ecosystem, MLIR dialects, or equality saturation frameworks (egg, eqsat)
• Experience with autotuning or cost modeling for performance optimization
• Knowledge of probabilistic algorithms and SMT solvers (Z3)
• Familiarity with tensor compiler frameworks such as Mirage, Halide, TVM, Triton
• Publications or projects in compilers or program synthesis
• Experience with workload optimization for Intel architectures (AMX, AVX-512, Sycl)

• Explore design and implementation of a PEG (Graph + PEG) abstraction combining algebraic optimization and hierarchical scheduling
• Apply E-graph-based equality saturation to explore equivalent operator compositions
• Develop multi-level schedule representations including loop tiling, vectorization, and memory placement focusing on Xeon CPU and Intel GPU targets
• Implement cost-driven and constraint pruning using resource-aware models and constraint satisfiability evaluation
• Integrate with MLIR's retargetable backend infrastructure for multi-target code generation
• Prototype equivalence checking using probabilistic and symbolic methods

All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, national origin, ancestry, age, disability, medical condition, genetic information, military and veteran status, marital status, pregnancy, gender identity or expression, sexual orientation, or any other characteristic protected by local law. This role is eligible for a hybrid work model allowing split time between on-site and off-site work.