Hire Ekambaranath T. - Machine Learning Engineer

Working in the AI Frameworks team to enhance PyTorch support for Intel GPUs, improving performance, stability, and compatibility for next-gen GPU architectures. Enabling high-throughput LLM inference frameworks (vLLM, SGLang) on Intel GPUs, focused on models such as Llama, DeepSeek, and GPT-OSS, optimizing runtime efficiency and deployment readiness. Contributed to the MLOps workflow for Intel Edge AI Services, supporting model packaging, deployment automation, and reproducibility across environments. Developed and deployed LLM inference pipelines on resource-constrained Intel Edge devices, leveraging OpenVINO + Hugging Face, focusing on latency and memory efficiency. Worked on integrating OpenVINO with TensorFlow, enabling optimized inference paths and improving framework-level compatibility for edge deployment.

* Developed AI/ML test cases using Python and RAILS, executing vast amount of structured test scenarios across CV and ML systems, leveraging OpenVINO, NPU/VPU, and MEP frameworks for hardware-accelerated inference. * Proactively resolved critical ML & CV issues, improving inference reliability, accuracy, and model stability on VPU/NPU hardware. * Monitored and analyzed nightly test runs, automated failure detection, and improved operational efficiency across heterogeneous datasets.

Worked with Appian technology to design and develop a Loan Process Management project, automating and streamlining the entire loan application workflow, improving operational efficiency for both consumers and the organization. The project enhanced customer experience by reducing loan processing time by 30%, providing a seamless, faster loan approval process and offering real-time tracking for consumers.