MangoBoost is looking for a Software Engineer to join the LLMBoost team to build, optimize, and scale AI workloads for high-performance, multi-node environments on both AMD and NVIDIA GPUs.
Requirements
- Proficiency in Python and Linux-based development environments
- Can squeeze extra performance out of GPUs through kernel tuning and advanced profiling techniques
- Know your way around Kubernetes and Docker for managing large-scale, containerized AI workloads
- Have hands-on experience running multi-node GPU training and inference jobs
- Understand the ins and outs of MLPerf or other AI benchmarking suites
- Can spot and fix network or I/O bottlenecks in high-throughput AI pipelines
- Have contributed to open-source AI/ML frameworks and love improving the tools the community relies on
Responsibilities
- Develop, optimize, and test features for the LLMBoost platform, including model deployment, auto-tuning, and multi-GPU orchestration
- Contribute to performance benchmarking and optimization on both AMD and NVIDIA hardware
- Maintain infrastructure for automated builds, testing, and releases
- Collaborate with the team on debugging, profiling, and improving system reliability
Other
- BS/MS/PhD in Computer Science, Electrical Engineering, or a related field (or equivalent experience)
- Strong problem-solving skills and eagerness to work on large-scale AI systems
- If you're creative and independent, with a genuine passion for technology, we want to hear from you.
