After Agents: The Next Chapter for Multimodal Products

What happens when generation models stop being tools and become infrastructure? Image and video generation used to be something humans triggered manually. Now it is increasingly called by agents inside real products.

From marketing tools generating product videos automatically to creative platforms building multimodal workflows, generation models are becoming part of product infrastructure.

This evening brings together founders, builders, and product teams working on multimodal AI. Hear directly from the Seedance and Seedream teams at ByteDance, join a panel with founders building generation into real products, explore live demos, and connect with other builders working on multimodal AI.

We’ll close the evening with networking, drinks, and a whole roasted lamb dinner.

Startups: apply to receive up to $2,000 in Seedance credits to experiment with the latest multimodal models.

What We'll Explore

What new multimodal models make possible now. Seedance and Seedream bring major improvements in controllability, speed, and generation quality. What kinds of products become possible when these models are reliable enough for real workflows?
What breaks when generation moves into real product workflows. Content, marketing, e-commerce, design tools. Who's shipping, what's working, and where it still breaks.
When agents become the caller. In many new products, generation models are no longer triggered by humans but by agents inside workflows. That changes API design, orchestration, latency requirements, and even what quality means in production.

Who Should Attend

Founders and product leaders building multimodal AI products.
Engineers working on generation APIs, multimodal pipelines, or agent workflows.
Creators and teams using AI generation in real production workflows.

Speakers

Natalie Hou, Enterprise AI Partnerships @ Bytedance
- Sharing what's new in Seedance (video) and Seedream (image) — the generation models behind ByteDance's products, and what they now make possible for builders.
Nico Maffey, Founder @ Sequencer Media
- Building an AI filmmaking platform where multiple generation models are orchestrated through node-based agentic workflows for video, image, and audio production.
Nick Pezzotti, Founder @ Trimz
- Former product lead for Veo (video generation) and Imagen (image generation) at Google DeepMind, now building a platform for AI-generated video content.
Rissa Cao, Co-Founder & CEO @ Fish Audio
- Previously co-founded PixAI, scaling it to 10M+ ARR and 100M+ images generated per month. Now building voice generation infrastructure at Fish Audio, serving 6M+ creators with APIs increasingly called by agent workflows.

Agenda

4:30 PM Doors open, check-in & early networking

5:00 PM Opening remarks

5:10 PM Keynote: Insights from Seedance & Seedream Models

5:30 PM Panel: Multimodal Products After Agents How generation models are moving from standalone tools into real product infrastructure.

6:10 PM Networking, Seedance demo area, whole roasted lamb dinner & drinks

About our community: Seamate

Seamate is a community for founders, builders, and investors exploring AI, startups, and global markets. With 15,000+ members across 10+ countries, we connect entrepreneurs across the U.S., China, and Southeast Asia and beyond through events, conversations, and collaborations.

Multimodal AI x Agents: Founders & Builders Night