We are excited to share that Echo AI has been officially accepted into the NVIDIA Inception program, NVIDIA's global community for startups defining the future of AI and accelerated computing. We join thousands of teams worldwide building on NVIDIA's stack, with direct access to the resources, expertise and partners that make ambitious AI products possible.
What is NVIDIA Inception?
NVIDIA Inception is a free program designed to help AI and data science startups build faster and grow with NVIDIA's technology. Selected companies get cloud credits, preferred pricing on hardware and software, technical training, go-to-market support and curated introductions to investors and partners.
What this unlocks for Echo AI
Cloud credits from NVIDIA Cloud Partners to experiment with larger models and heavier inference workloads
Preferred pricing on GPUs and NVIDIA software to scale our infrastructure as our user base grows
Access to NVIDIA Nemotron open-weight models for evaluating specialized agent capabilities alongside our existing OpenAI stack
NVIDIA Dynamo for studying multi-node inference serving as Echo handles more concurrent conversations across channels
NVIDIA NeMo for building, monitoring and optimizing agent lifecycles at scale
NVIDIA Developer Program with SDKs, technical documentation and training
VC connections and partner offers through the Inception network
What this means for our roadmap
To be transparent: we do not ship any feature built on NVIDIA infrastructure today. Every Echo currently runs on OpenAI for chat and voice, and ElevenLabs for custom voice. What Inception gives us is the runway to seriously evaluate the next layer of the platform without that work blocking our day to day operations.
Here is what we are exploring now that we have the resources to do it properly:
Lower latency for voice and chat by benchmarking accelerated inference paths against our current stack
Specialized mission agents using open-weight models for narrow tasks like classification, routing and on-platform safety checks, while keeping OpenAI for the main conversational layer
Heavier multimodal workloads including faster image generation and longer video understanding for media-rich Echos
Better observability for agent runs, traces and quality scoring as Echos handle more high-stakes commerce and booking flows
Infrastructure resilience with a clearer fallback story if a primary provider has an incident
Why this matters
The economics and physics of AI are still moving fast. Being part of Inception means we get to test what is genuinely better for our users, not just what is most convenient. Every decision we make on models, latency and cost flows directly into pricing, reliability and the kinds of Echos businesses can realistically run.
A big thank you to the NVIDIA Inception team for the welcome. We will keep our principle simple: ship what is measurably better for the businesses building on Echo, and stay honest about everything still in evaluation.
What is next?
We will share concrete updates as soon as any of this graduates from evaluation into a shipped feature. In the meantime, expect the usual cadence of product updates, with the same OpenAI-powered chat and ElevenLabs-voice you already know.
Learn more about the program at nvidia.com/startups