Assist in the design, development, implementation and deployment of Gen AI and LLM applications
Contribute towards implementing fine tuning and distillation techniques
Apply Systems knowledge and experience to research, design, develop, simulate, and/or validate systems-level software, AI hardware, architecture, deep learning algorithms, and AI solutions while ensuring system-level requirements and standards are met
Perform AI model benchmarking and functional analysis to drive requirements and specifications
Propose deployment strategies with AI model/workload optimization and deployment
Build AI solutions and develop and analyze system level design including requirements, interface definition, functional/performance definition, and implementation of a new system or modification of an existing system
Collaborate with own team and other teams to complete project work, including implementing and testing features and verifying the accuracy of AI systems
Keep abreast with the latest advancements in the AI/ML space (models, HW/SW) and drive innovative solutions.
Develops new and innovative ideas for a product or feature area
Drives triage of problems at the system level to determine root cause and presents results of testing and debugging to team members.
Required Skills
Strong proficiency in Python, and ML frameworks (PyTorch, TensorFlow).
Deep understanding of ML development, deployment and applications.
Deep understanding of system performance profiling and parallel computing.
Preferred Qualifications
Master's or PhD Degree in Engineering, Information Systems, Computer Science, Physics or related field
Good understanding of GenAI architectures from transformers, diffusion, hybrid - LLMs, LVMs, embeddings
Working experience with fine-tuning GenAI models and Reinforcement Learning
Background in compiler optimizations for ML workloads is a plus
Experience with architectural Patterns for Large-Scale AI Systems: Knowledge of microservices, and distributed systems
Optimize inference performance across heterogenous nodes CPUs, GPUs, and specialized accelerators
Experience in implementing caching, batching, and parallelization strategies for high-throughput systems.
Familiarity with hardware acceleration
Proficiency with version control systems (Git) and code review tools (Gerrit, GitHub, GitLab) and collaborative development workflows