Ali

Artificial Intelligence Engineer

Mercurial Minds • Full-time

Sep 2022 - Present • 3 yrs 8 mos

Artificial Intelligence Engineer — Jazz Tamasha (Current) Built and deployed large language model (LLM) services using vLLM and Kubernetes for scalable AI inference. Implemented GPU optimization and KV caching to improve response speed and reduce inference cost. Designed AI pipelines for speech-to-text and translation, benchmarking models like Whisper v3 and Seamless M4T for production use. Integrated RabbitMQ and Celery task queues to handle high-throughput AI workloads in production systems. Developed AI-powered automation tools for transcription, translation, and data processing. Key Achievements Optimized LLM inference pipelines improving throughput and latency for real-time AI applications. Benchmarked multiple AI models and produced technical performance reports with metrics and charts. Built production-ready AI microservices deployed on GPU infrastructure.