Builders 👋
The @ChronoSeek MCP Server is now live.
🔗 https://t.co/k1ZPB7n8ch
Available tool groups:
• 📚 Bittensor knowledge retrieval
• 🔍 Subnet repository analysis
• 📈 Live liquidity & alphanomics reports
• 📊 Subnet performance summaries
The goal is to make it easier for AI agents and developer tools to interact with the Bittensor ecosystem through a standard MCP interface.
This is only the first release - many more tools are on the roadmap.
If you’re building on #Bittensor and ChronoSeek, we’d love to hear what capabilities you’d like us to add next 🚀
We’ve updated the Subnets page to reflect the latest chain changes, so metrics such as Users Held, Alpha in Reserve, Deregistration Price, and other subnet statistics now match the current network state.
We’re also experimenting with alternative indexers to speed up on-chain storage scanning and make the dashboard even more responsive.
More improvements coming soon 🚀
We’ve just opened up one of the internal platforms we’ve been using to research the Bittensor ecosystem.
🔬 https://t.co/p8zVIQZdon
It’s an experimental platform featuring tools for:
• Subnet analytics
• Mempool insights
• Network research
• And more to come
The goal is simple: build tools that are genuinely useful for the community.
We’re actively looking for feedback:
• What would you like to see next?
• What challenges do you face day to day?
• What tools would make your workflow easier?
Have a look, let us know your thoughts, and help shape what we build next. 🚀
Our 2nd subnet integration is now live 🎉
We’ve integrated @vidaio_ into @ChronoSeek as part of our hardened synthetic task generation pipeline.
Validators now use VidAIo API to compress synthetic task videos before they’re distributed to miners, adding another layer of transform diversity and making benchmark fingerprinting and exploitation more difficult.
This marks our second subnet-to-subnet integration on Bittensor, following our recent Hippius integration.
A big thank you to the VidAIo team for the great work and support.
We’re continuing to build an interconnected Bittensor ecosystem, with more subnet integrations on the way 🚀
@ChronoSeek is now integrated with @hippius_subnet
This is our first subnet-to-subnet integration on @bittensor, with more on the way.
Why this matters:
Instead of exposing original dataset URLs, validators now upload transformed synthetic task videos to Hippius. Miners only receive the Hippius-hosted artefacts, making it significantly harder to infer where tasks originated from.
This adds another layer to our anti-gaming pipeline while keeping the miner-facing API unchanged.
A huge thank you to @mast3rdubs@mogmachine for their support throughout the integration.
Next up: @chutes_ai and @vidaio_ . We’re excited to continue building a more connected Bittensor ecosystem, one integration at a time. 🚀
@ChronoSeek is now integrated with @hippius_subnet
This is our first subnet-to-subnet integration on @bittensor, with more on the way.
Why this matters:
Instead of exposing original dataset URLs, validators now upload transformed synthetic task videos to Hippius. Miners only receive the Hippius-hosted artefacts, making it significantly harder to infer where tasks originated from.
This adds another layer to our anti-gaming pipeline while keeping the miner-facing API unchanged.
A huge thank you to @mast3rdubs@mogmachine for their support throughout the integration.
Next up: @chutes_ai and @vidaio_ . We’re excited to continue building a more connected Bittensor ecosystem, one integration at a time. 🚀
Major validation upgrade for ChronoSeek🔒
We’ve implemented a new anti-gaming framework focused on:
• Private task artifacts
• Query paraphrasing
• Secret-seeded sampling
• Hard negatives & canaries
• Transform diversity
• Enhanced telemetry
The miner-facing API remains unchanged.
The goal: reward genuine video understanding, not benchmark memorisation.
A big step towards production-grade evaluation on our Video Moment Retrieval subnet on #Bittensor $TAO 🚀
🤖 ChronoGuide Update — Now Metagraph-Aware
ChronoGuide has received a major upgrade.
Previously, it was primarily focused on helping newcomers understand:
• What ChronoSeek is
• How the subnet works
• General project onboarding
Now, ChronoGuide is fully metagraph-aware.
This means it can now assist with:
• Miner-related questions
• Validator-related questions
• Subnet operations
• Network state & participant awareness
• More advanced ecosystem discussions
The goal is to evolve ChronoGuide from a simple onboarding assistant into a true AI-powered subnet companion for the community.
You can still:
→ Tag @ChronoGuide directly (on Discord)
→ Create dedicated threads for ongoing investigations/discussions
Would love feedback on:
- Accuracy
- Missing capabilities
- Useful operational features we should add next
More improvements coming soon 🚀
We’ve launched ChronoGuide 🤖
An AI-powered Discord assistant for the ChronoSeek ecosystem.
It can:
• Explain what ChronoSeek is
• Help setup miners & validators
• Answer community questions
• Guide users through APIs/docs
• Assist newcomers onboarding into the subnet
•••
Usage:
→ Tag @ChronoGuide in chat
→ Create threads for recurring AI-assisted discussions & investigations
This is part of a bigger vision:
AI-powered community infrastructure around decentralized intelligence networks on Bittensor.
Making the ecosystem more accessible, educational, and scalable 🚀
If you’re interested in:
AI × video × decentralized intelligence on Bittensor
Come join us 👇
We’ve launched ChronoGuide 🤖
An AI-powered Discord assistant for the ChronoSeek ecosystem.
It can:
• Explain what ChronoSeek is
• Help setup miners & validators
• Answer community questions
• Guide users through APIs/docs
• Assist newcomers onboarding into the subnet
•••
Usage:
→ Tag @ChronoGuide in chat
→ Create threads for recurring AI-assisted discussions & investigations
This is part of a bigger vision:
AI-powered community infrastructure around decentralized intelligence networks on Bittensor.
Making the ecosystem more accessible, educational, and scalable 🚀
If you’re interested in:
AI × video × decentralized intelligence on Bittensor
Come join us 👇
We've just launched the official ChronoSeek Discord 🎉
As we continue building our Video Moment Retrieval subnet on Bittensor, we want a place for:
• Builders & developers
• Early users
• Contributors & collaborators
If you're interested in semantic video search, AI infra, or building on top of ChronoSeek - come join us.
Let’s build this together 🚀
https://t.co/9KS6BI2Xug
@dsvfund We should connect 🤝
We already have everything prepared on our end, and the work has been verified by several well-known names in the community.
What’s the best way to reach you?
@markjeffrey@siam
Update: Miner & Validator Leaderboards are now live on the ChronoSeek dashboard 📊
You can now explore the metagraph and track the status of participants in our Video Moment Retrieval subnet on Bittensor.
→ Miner activity
→ Validator participation
→ Network visibility
🔗 https://t.co/uQAooNROgB
Next:
→ Deeper performance metrics
→ Scoring transparency
Building an open, inspectable intelligence network 🚀
New release: ChronoSeek Subnet Dashboard 📊
You can now track the current state of our subnet on Bittensor in real time.
→ Network activity
→ Subnet status
→ Protocol-level visibility
🔗 https://t.co/uQAooNROgB
This is a step toward making the system more transparent and measurable.
Next:
→ Detailed miner & validator views
→ Deeper performance insights
Building the foundation for an open intelligence network 🚀
Honored to have placed 2nd in the @HackQuest_ hackathon with ChronoSeek 🥈
Huge thanks to the organizers, judges, and the community for the opportunity and support throughout.
We built a Video Moment Retrieval subnet on Bittensor — from idea → working MVP (demo + API) during the hackathon.
This is just the beginning. Excited to keep pushing this forward 🚀
Bittensor Subnet Hackathon Winners Announced 🏆
After a strong Round 2 of testnet builds, live demos, and subnet execution on @opentensor, here are the final Top 7 teams:
🥇 Proven — Christopher Hardy Gunawan
🥈 Chronoseek — Connor Daly
• Mentiss_AI — Jeremy Wang
• Sotarad — Wade Walton
• Defektr — Nejc Rožman
• OpenMind — Bello Iteoluwakisi
• BitDefense — Aleksei Gubin
Honorable Mentions:
• Moirai — Kael Reed
• Talos — Roderich Cavine Chow
• Vividverse — Daniel Braine
• Query Agent — Daniel Derefaka
• TensorClock — Valeriy Lihachev
• C-SWON — Aditya Singh
Huge congratulations to all builders who joined and shipped ideas throughout our hackathon 🔥
Check out the Top 7 projects below 👇
ChronoSeek just got a major upgrade 🔊
We’ve added audio understanding (transcription-based) to ChronoSeek on Bittensor.
Now:
→ Queries can match against spoken content
→ Better retrieval for dialogue-heavy scenes
Under the hood:
• New transcription pipeline
• Increased timeouts to support audio processing
Tradeoff (for now):
→ Slightly higher latency — worth it for better results
Next step:
→ Beyond speech → full sound understanding
Multimodal retrieval is coming together 🚀
LLMs understand text.
Some models understand images.
But almost no systems truly understand video.
Because video isn’t just pixels — it’s sound.
ChronoSeek is changing that.
We’re evolving our subnet on Bittensor from visual-only → multimodal understanding.
Phase 1:
→ Speech-aware retrieval (via transcription)
Phase 2:
→ Full sound understanding (not just words — real-world audio)
This is how you make video actually searchable.
Not clips. Not captions.
Moments.
We’re building the infrastructure for it 🚀