Role Purpose 

We are seeking a result-oriented Software Developer with expertise in Generative AI & Large Language Models to design & develop solutions powered by Fujitsu’s new processor, which help solve real-world challenges facing society and businesses across different industries. The role involves building and optimizing LLM, LVM & LMM. The ideal candidate will have knowledge, experience, and passion for both AI models/applications and hardware/software system & should be skilled in back-end development, high-performance computing, AI framework engineering on modern CPU-based architectures. 

Responsibilities 

• Improve and analyze performance of Generative AI & LLM software applications. 

• Enable and optimize OSS/ISV applications for Fujitsu’s new processor, starting with LLM-related applications. 

• Develop new algorithms for ML/DL frameworks, tuning technologies and working on software based on the proposed approaches using AI framework engineering. 

• Improve and analyze performance of Generative AI & LLM Models for inferencing and fine tuning. 

• Understanding of AI model optimization techniques such as operator fusion, quantization & Sparsity. 

• Solid understanding of hardware & software systems 

• Troubleshooting, debugging, and fixing bugs and upgrading software/applications. 

• Writing technical documentation 

• Learning continually, sharing knowledge, and fostering exchange of skills 

• Working using agile methods (planning meetings, review meetings, standup meetings, development, etc.) 

• Working on multiple projects at once while keeping focused on project timeline 

Key Performance Indicators 

• Number of software applications enabled and optimized for Fujitsu’s new processor. 

• Number of software applications with improved performance 

• Development of new software technologies 

Experience 

You will be able to demonstrate that you have: 

• A degree in Computer Science, Engineering, or a related field (Master’s/MTech or PhD preferred) 

• Minimum 5 years’ experience in SW development, Deep Learning & AI Frameworks. 

• Proficiency with fundamental back-end server-side languages such as Python & C/C++. 

• Knowledge in neural compression & quantization techniques such as pruning & knowledge distillation 

• Knowledge of LLM Fine tuning & training techniques (PEFT, LoRA, QAT etc. are preferred) 

• Knowledge and experience related to AI Agents, RAG & RLHF Techniques. 

• Familiarity with Inference runtimes ONNX Runtime, OpenVINO. 

• Experience with Deep learning frameworks such as PyTorch, TensorFlow, JAX, Llama.cp & vLLM. 

• Experience in software development in agile approach 

• Strong debugging skills using profiling tools (PyTorch Profiler, TensorFlow Profiler, perf). 

• Excellent writing, verbal communication, and teamwork skills 

Preferred Experience: 

You will be able to demonstrate that you have: 

• Knowledge of hardware-aware optimizations (oneDNN, XLA, ROCm, MLIR, SparseML). 

• Contributions to open-source community 

• Publications in International forums / conferences / journals 

• Knowledge and experience in Cloud service (Azure/AWS) features such as Functions, VM, Container, DevOps (CI/CD) 

• Great skills in evaluating performance and security of software applications and delivering solutions which are efficient and performant.