Stacking Alpha
Joined February 2025
- Tweets468
- Following226
- Followers1.2K
- Likes624
Alpha | Markets · AI · Research retweeted
Incredibly excited for this! Ninja was a good experiment into incentivizing harnesses, but ultimately we found the problem was too intractable for Bittensor. The team and I believe this problem we are now solving is among the most important in Bittensor!
Alpha | Markets · AI · Research retweeted
The Bittensor $TAO ecosystem weekly update Vol.13
This was a big week, for us personally but also for the ecosystem.
A 🧵
Alpha | Markets · AI · Research retweeted
I've been thinking a lot recently about what's actually been holding Bittensor back from the rest of crypto. It isn't the tech. It isn't the teams. It certainly isn't ambition.
It's that most of crypto still doesn't really see what's being built here.
That's about to change.
This week, @taodotcom is opening a Solana-native route into TAO. And the first people I'm hoping show up are Solana traders.
Solana traders are some of the sharpest, fastest, most aggressive capital allocators in crypto. They've taken enormous bets on projects with anonymous teams, unaudited tech, and sometimes on nothing more than a narrative. That isn't a criticism. That's just the market doing its job, on the chain that moves the fastest.
Now point that same energy at Bittensor.
Doxxed teams. Real AI infrastructure. Subnet businesses with product, users, and roadmaps. An ecosystem that has spent years in the trenches building the thing instead of selling it. Talk to anyone serious in this space and you'll see the same pattern. These are operators, not narrative merchants.
That's what @taodotcom is opening up.
Better access. Better trading conditions. A clean venue where Solana-native capital can take a real position on one of the most substantive AI ecosystems in crypto, without ever having to leave the chain they already live on.
What it brings in is liquidity, capital, fresh eyes, and a lot of new ideas about how this ecosystem can keep growing.
For $TAO holders and subnet ecosystems, this is a new liquidity source and a new demand surface. Solana capital meeting Bittensor product. Two ecosystems lined up on the one thing that actually matters in crypto, which is building large businesses on chain.
Solana is crypto-native. Bittensor sits at the intersection of crypto and AI. There's a lot we can learn from each other, and the overlap is where this is going to get really interesting.
Honestly, in this ecosystem the people who show up early are almost always the same people who end up building the bridges.
No one is bigger than Bittensor. But Bittensor gets bigger when the rest of crypto can see it.
Alpha | Markets · AI · Research retweeted
Green Compute is built to turn real enterprise demand into subnet rewards. Enterprises buy inference from us, demand is then routed through validators and miners earn by supplying the compute that serves it.
What makes SN110 different is the focus on enterprise-grade symmetry across hardware, networking, IPs, and engineering support. So larger contracts can be distributed reliably across multiple miners, instead of relying on fragmented capacity.
Alpha | Markets · AI · Research retweeted
First announcement of April.
We introduce Quasar-3B (1B Active) a looped continuous-time transformer built for long-context intelligence
Quasar uses a hybrid architecture combining Quasar layers + GLA, enabling efficient stateful reasoning with stable long-range dependency handling.
This release is the base model of the Quasar system, designed for distributed training and distillation at scale. In the coming days, SN24 miners will begin working with us to distill knowledge from frontier models like Qwen into Quasar, pushing toward a new SOTA in long-context modeling.
The training roadmap is staged:
Stage 1: Quasar-RoPE stable pretraining with 16K context, establishing the base representation for the system
Stage 2: Quasar-DroPE continued training with distillation as the core mechanism, removing positional encodings and scaling toward 5M token context
In both stages, distillation remains the central training signal, driven collaboratively through SN24 miners using knowledge transfer from frontier models.
Read more about Quasar and get ready for mining 👇
Alpha | Markets · AI · Research retweeted
IT’S COMING HOME 🏆
Out of 1000+ startups that applied to the @signalweek startup competition, we take the win.
Big moment for us and for the Bittensor eco. Huge congrats to @arnod3f and @manakoarchie on the stage today.
Alpha | Markets · AI · Research retweeted
If you are new to $TAO and you want to understand it better, do watch this interview with Const, the co-founder of Bittensor.
He explains how Bittensor works but also goes into his broader view on crypto which all translates in his idealism and why he build bittensor.
I found it really really interesting, Const is a very good speaker and threadguy seems to be genuinly curious (which is perfect).
Const joins from minute 28 to 1:27.
IM LIVE: Trump's TACOs Aren't Working... - IRAN vs USA Continues, TAO is RIPPING, Crypto is BACK!? Interviewing @const_reborn (4pm et)
Twitch.tv/threadguy nitter.cf/i/broadcasts/1nJOLEnDg…
Alpha | Markets · AI · Research retweeted
Ridges agents aren’t assistants.
They take ownership.
Here’s what makes them autonomous ↓
Alpha | Markets · AI · Research retweeted
This is getting exciting!
Michael Taolor(@michaeltaolor) advocating for Bittensor adoption as the currency of the agent economy.
-First Post on moltbook.
-First Wallet generated.
-Next...
Go Taolor, on board those agents!
I'm claiming my AI agent "MichaelTaolor" on @moltbook 🦞
Verification: swim-HNUL
Alpha | Markets · AI · Research retweeted
Capitulation Watch: Three Signs the Washout Is Getting Real
1. Bitcoin implied volatility (BVIV) at 88.55, closing in on the FTX-collapse peak of 105 (h/t @volmexfinance)
2. 8th largest Coinbase trading day ever by USD value ($3.34B). At ~$62K, that's roughly 54,000 BTC changing hands in a single session. In unit terms, this likely ranks even higher than the USD figure suggests, since many of the top USD volume days occurred at significantly higher prices.
3. Daily RSI at 15.64, at or below March 2020 COVID crash lows
Margin calls are firing. Forced liquidations are likely still working through the system. This has the signature of a capitulation event, but capitulation can be a process, not a single candle (unless we get a massive wick!)
These conditions can persist for weeks or even months before a durable low forms.
Alpha | Markets · AI · Research retweeted
🚨 PRODUCT UPDATE – Ridges SN62 is going live
Open Beta on Feb 19
– No waitlist. Full access.
– Create a task. An agent picks it up. PR delivered.
– Code ingestion and end-to-end execution
– Multilingual support: Python and beyond
Incentive Mechanism Overhaul
– Manual approvals eliminated
– New system reduces score variance
– Native antifraud and overfit detection
– Higher agent throughput per day
Enterprise-Ready Stack
– 100% open source infra, models, and agents
– On-prem deployment for compliance
– TEE support for off-prem secure compute
The Next Frontier
– From benchmarks to real product usage
– Evaluation loop tied to user behavior
– Built to become a daily driver
– Co-dev, audit, refactor — all in one stack
The product is ready. The market is next.
Now it’s your move: Buy, buy more or don’t buy?
The $TAO Bittensor ecosystem is definitely non-stop building.
Source: Live X Spaces AMA with the Ridges' team: @ridges_ai @KibibyteMe @hobbleabbas @latentholdings
Alpha | Markets · AI · Research retweeted
Loosh Inference Subnet - Now Live on Testnet!
We're excited to announce the official launch of the Loosh Inference Subnet (Subnet 78) on Bittensor! Our code repositories are now public, testnet is active, and we're inviting miners and validators to join us.
What is the Loosh Inference Subnet?
The Loosh Inference Subnet is a decentralized, high-quality LLM inference network that powers real-time AI agents and applications. We're building the inference infrastructure for the next generation of agentic AI systems, including the Loosh AI agent at app.loosh.ai
How It Works
Our subnet uses a sophisticated multi-layered architecture:
1. **Users/Agents** send inference requests to our Gateway
2. **Gateway** creates challenges and stores them in the Challenge API
3. **Challenge API** publishes challenges to validators via Fluvio streaming
4. **Validators** distribute challenges to multiple miners simultaneously using **Fiber MLTS encryption**
5. **Miners** perform LLM inference and respond with encrypted answers
6. **Validators** evaluate responses using consensus-based scoring, embeddings, and quality metrics
7. **Best response** is selected and returned to the user in real-time
**Key Differentiators:**
- **Real Production Use** - Powers live AI agents at app.loosh.ai
- **Consensus-Based Evaluation** - Quality over speed, ensuring high-quality responses
- **Secure Communication** - Fiber MLTS (Multi-Layer Transport Security) for all miner-validator communication
- **Model Agnostic** - Use any model that works for your hardware (OpenAI API compatible format)
- **GPU Optimized** - Efficient deployment options for all hardware tiers
Alpha | Markets · AI · Research retweeted
Hi everybody,
Somehow, MEV bots are hitting Bittensor again.
Our "Neuralteq MEV protection" still works, so make sure to keep the correct settings like this:
- Slippage protection on
- Max slippage 0.25
- Allow partial orders
- Neuralteq MEV protection turned on
Sign your tx within 10 seconds and enjoy swapping. dashboard.neuralteq.com/subn…
$dTAO $TAO #Bittensor
