The Developer - I/O Acceleration role at IBM focuses on optimizing the data path from disk to GPU memory for the Gala Lakehouse analytics platform. The I/O Acceleration team is responsible for enhancing performance through accelerated I/O and decompression techniques. This position requires strong expertise in modern C++ and Linux systems to improve throughput and reduce latency in data-intensive workloads. The successful candidate will work collaboratively with various teams to design effective APIs for accelerated I/O.

• High School Diploma/GED is required, while a Bachelor's Degree is preferred.
• Strong modern C++ programming skills and deep understanding of Linux systems internals are essential.
• Hands-on experience with storage I/O subsystems, decompression, or query-engine data paths is required.
• Familiarity with GPU-aware pipelines or adjacent acceleration frameworks like CUDA or GPUDirect is necessary.
• Experience in delivering production software in Agile environments is required.

• Design, build, and optimize accelerated I/O and decompression paths for data-intensive analytics workloads.
• Improve end-to-end throughput across the storage, network, host, and GPU boundary.
• Integrate with GPU-aware runtimes and high-bandwidth fabrics, tuning for Blackwell-class hardware.
• Build benchmarks and microbenchmarks to identify I/O cliffs, queue contention, and tail latency.
• Collaborate with query engine, storage, and hardware teams to co-design usable APIs for accelerated I/O.

• IBM offers competitive compensation, healthcare coverage, retirement programs, paid parental leave, tuition assistance, wellness programs, flexible work options, and extensive learning and certification resources.