Bhavishya Mittal

Architected and scaled production ML pipelines and RAG systems, reducing model deployment cycle time by 35% and ensuring 99.9% uptime while improving response accuracy by 20% and cutting hallucinations by 15%. Built distributed backend and inference infrastructure to handle 100M+ requests/month and reduced query latency by 40%.

Directed architecture and deployment of global ML systems ensuring 99.99% reliability and low-latency inference, and pioneered LLM/RAG solutions that improved personalization and relevance by 25%. Optimized training and inference pipelines to cut training time by 30% and inference latency by 25% while reducing compute costs by 20%.

Designed and developed backend services and distributed systems for enterprise-scale ML workloads, improving model inference throughput and reducing latency by 15%. Built ETL and analytics workflows to accelerate feature engineering and support petabyte-scale datasets.

Developed and optimized internal software tools to enhance engineering workflow efficiency and contributed to feature design and implementation with a focus on performance and scalability. Participated in code reviews and testing to maintain high-quality internal deployments.

Implemented core application features for enterprise products in a C#/.NET and SQL Server environment, improving performance and maintainability through query optimization and code refactoring. Collaborated with global engineering teams to deliver aligned product features.

Developed and maintained web application components using PHP, MySQL, and JavaScript, delivering new features and bug fixes to improve system functionality and client satisfaction. Collaborated with a small team to translate client requirements into efficient solutions.

Earned a Master of Science in Computer Science from Georgia Institute of Technology (2015–2017).

Completed a Bachelor of Technology in Computer Science at Indian Institute of Technology Kanpur (2011–2015).