Multimodal Weekly 84: Accelerating Multimodal Model Development with Twelve Labs and SuperAnnotate

​​​​​​In the 84th session of Multimodal Weekly, we will explore how Twelve Labs and SuperAnnotate work together to streamline the process of fine-tuning high-performing multimodal models.

​SuperAnnotate provides a flexible platform for both data preparation and qualitative evaluation, where expert review, custom workflows, and high-quality annotations play a critical role.

Abstract

​We'll walk through:

• ​A typical Twelve Labs fine-tuning workflow, and where SuperAnnotate fits in during data labeling and evaluation.

• ​How teams can build custom annotation workflows in SuperAnnotate that output in the Twelve Labs fine-tuning format, using orchestration templates to push data seamlessly into the training pipeline.

• ​Real-world examples of multimodal use cases supported in SuperAnnotate, such as:

  • ​Video-to-text caption generation and summary evaluation

  • ​Video Segmentation

  • ​Multimodal agent scoring where human reviewers assess output reasoning or relevance

​SuperAnnotate helps teams move faster by giving them tools to customize annotation UIs, manage SME reviewers, and generate structured feedback on model performance—ultimately helping Twelve Labs customers train better models, faster.

Speakers

✅ Adam Smith is the Head of Partnerships at SuperAnnotate, where he works with customers, model providers and ML platform companies to drive multimodal AI projects through high quality human in the loop workflows. He previously was Head of Partnerships at Jasper AI and also spent 16 years at Google leading GTM and partnerships for a variety of enterprise product businesses including Vertex AI focusing on NLP and computer vision. Prior to that he built and led a team managing Google Cloud’s top ecosystem partners.   He is also a founding team member of Google Workspace and Chromebooks.

✅ Vignesh Sridhar is a Solutions Engineer focused on partnerships, where he focuses on integrating SuperAnnotate’s capabilities with partner environments to build GTM solutions. He recently graduated from Carnegie Mellon with a masters degree and prior to that was working as a solutions architect at Dell designing data center infrastructure solutions.

​Join the Multimodal Minds community to connect with the speakers!

Multimodal Weekly is organized by Twelve Labs, a startup building multimodal foundation models for video understanding. Learn more about Twelve Labs here: https://twelvelabs.io/