hi, i'm nikitha

i lead product for inference at together.ai in san francisco, helping teams run open-weight and custom models in production.

i launched provisioned throughput to give teams reserved inference capacity with token-based pricing, and led the next generation of our production inference platform to make deploying, scaling, and updating models simpler. my focus is helping teams balance latency, throughput, reliability, and cost as they build AI products at scale.

i also invest in cloud and compute infrastructure, AI/ML, developer tools, and data.