ICE
Indexless Context Engine. Turn-granular memory for AI agents.
Our first product
Explore ChapterDopove is a systems thinking company. We build useful memory for AI agents, and we apply the same thinking to how we run our own marketing.
Dopove was founded by a technical builder who applies systems thinking to both product design and company operations, rather than separating technical excellence from marketing chaos.
Our current exploration is applying systems thinking to reduce founder marketing workload through an autonomous marketing approach. We build in public: we share what works, what does not, and what we are learning, about agent memory and about running a better company.
ICE keeps the order of events, retrieves by lane, and carries memory across sessions
GoingMerry, our next product, runs models on your own hardware
Cortex is an open-source runtime for multi-agent systems
Four principles that guide every architectural decision
We ship fewer things better. Every API and every CLI command is intentional.
If it takes more than 3 lines to integrate, we failed. The best infrastructure is invisible.
No toy prototypes. Type-safe, auditable, monitored, and scalable from the first commit.
Cortex is open source on GitHub and the ICE Community Edition is free to run. Community drives adoption; enterprise drives sustainability.
GoingMerry is the next product we are building after ICE.
From founding to production-grade infrastructure
Dopove Private Limited founded in Chennai, India.
Our first product, the Indexless Context Engine.
Accepted into NVIDIA Inception program for AI startups
High-performance multi-agent runtime released on GitHub
Cross-agent memory layer for vibe coding workflows

Dopove is a member of the NVIDIA Inception program, which nurtures startups transforming industries with advancements in AI and data sciences.
© 2025 NVIDIA, the NVIDIA logo, and NVIDIA Inception are trademarks and/or registered trademarks of NVIDIA Corporation in the U.S. and other countries.
We're looking for engineers who care about memory architecture, compiler optimization, and making AI infrastructure disappear.