Subnet 14
TAO Hash
Latent Holdings
Tao Hash allows Bitcoin miners to redirect hash rate, earning alpha tokens for staking or trading.

SN14 : TAOHash
| Subnet | Description | Category | Company |
|---|---|---|---|
| SN14 : TAOHash | Bitcoin mining | Mining | Latent Holdings |
Tao Hash operates as a mining ecosystem where Bitcoin miners can redirect their hash rate towards the Tao Hash subnet and mine alpha tokens instead of Bitcoin. The system incentivizes miners to contribute their hash rate, and in return, they receive alpha tokens. These tokens can be used for staking or traded. This concept mirrors traditional mining pools but with a focus on a new tokenomics structure.
Miners give up mining Bitcoin, and instead, based on the amount of Bitcoin hash rate they provide, they are rewarded with alpha tokens. The subnet’s goal is to create a fair system that allows liquidity to build over time while preventing manipulation or unfair advantages, especially during the early stages when a new subnet is born. By ensuring that early price volatility settles and no insider trading occurs, the system allows for a decentralized, fair mining model.
The TAO Hash system is a decentralized marketplace that allows Alpha tokens to be automatically exchanged for Bitcoin hashrate. It operates on two fundamental principles:
Value Exchange: Miners contribute Bitcoin hashrate to validator pools. In return, they receive Alpha token rewards proportional to the value of their contributions. Validators are responsible for evaluating this contribution and distributing rewards through on-chain weight updates.
Incentive Alignment: The system leverages market mechanisms by pricing hashrate based on real-time Bitcoin hash prices. This approach naturally incentivizes efficient hashrate allocation and promotes competitive miner behavior.
System Roles and Components
The TAOHash system includes several core participants, each with specific responsibilities:
Miners (BraiinsMiner): Provide hashrate by allocating mining resources across validator pools.
Validators (BraiinsValidator): Evaluate miner performance and set network weights accordingly.
External Pools: Execute the actual Bitcoin mining operations and provide real-time metrics.
High-Level System Architecture
The architecture integrates components from Bittensor, external Bitcoin mining infrastructure, and external data sources to ensure smooth operation. The main architectural components include:
TAOHash Subnet 14: The designated Bittensor subnet responsible for handling hashrate-based tasks.
Bittensor Blockchain and Metagraph: Used for subnet synchronization, hotkey registration, and on-chain weight management.
External Mining Infrastructure: Actual mining tasks are carried out via connected mining pools and hardware.
Market Data Sources: Hash price information is pulled from services like Braiins Insights and CoinGecko.
Storage Systems: Persist miner data, schedules, and configurations using either Redis-based or JSON-based backends.
Key System Components
Miners (BraiinsMiner)
Miners manage and direct hashrate across multiple validator pools using scheduling and allocation strategies. They function through several key subsystems:
- MiningScheduler: Manages timed transitions between different validator pools.
- Allocation Strategies: Supports different methods of distributing hashrate — including stake-based, equal-weighted, and multi-pool distribution algorithms.
- BraiinsProxyManager: Interfaces with mining hardware by updating proxy configurations to redirect mining output.
- Storage Layer: Stores pool configurations and scheduler settings using either Redis or JSON-based systems.
Validators (BraiinsValidator)
Validators assess the quality and impact of miner contributions. Their responsibilities include:
- BraiinsPoolAPI Integration: Pulls real-time statistics and worker data from external mining pools.
- Hash Price Integration: Fetches live Bitcoin hash prices to convert hashrate into value terms.
- Scoring Algorithms: Uses time-windowed metrics (typically 5-minute and 60-minute windows) to evaluate performance.
- Weight Updates: Adjusts miner weights on-chain based on the calculated performance scores.
Storage Systems
Two types of storage backends are supported:
Redis Storage (BaseRedisStorage)
- Used for high-performance caching.
- Includes TTL (Time-To-Live) functionality for expiring outdated entries.
JSON Storage (BaseJsonStorage)
- Used for simple, local storage.
- Appropriate for smaller or less resource-intensive deployments.
Both systems extend a common interface (BaseStorage) and allow uniform access to pool configuration and scheduling data.
Data Flow and Integration Points
Bittensor Network Integration
- Supports hotkey registration and subnet joining via the Bittensor library.
- Keeps validators synchronized through the metagraph.
- Handles on-chain weight updates to ensure accurate and fair reward distribution.
External Mining Pool Integration
- Collects real-time worker performance metrics using the BraiinsPoolAPI.
- Verifies hashrate output by cross-referencing with mining pool data.
- Manages mining pool configurations through proxies to control actual mining behavior.
Market Data Integration
- Pulls live Bitcoin hash price data from APIs like CoinGecko and Braiins Insights.
- Uses this data to evaluate and price hashrate contributions fairly.
- Supports continuous incentive calibration based on current market conditions.
Network Parameters
The TAOHash system is configured with specific network parameters designed to align with Bitcoin’s mining and block generation rhythms:
Subnet ID: 14
(Specifies the Bittensor subnet under which TAOHash operates.)
Evaluation Window: 720 blocks (approx. 2.5 hours)
(Defines the time period used by validators to evaluate miner performance.)
Metrics Sampling Frequency: Every 5 minutes (25 blocks)
(Controls how often miner hashrate is measured and evaluated.)
Minimum Stake for Validators: 10,000 TAO
(The minimum amount of TAO required to run a validator on this subnet.)
Minimum Block Allocation per Validator: 40 blocks
(Defines the shortest allocation window to ensure reliable hashrate measurements.)
These parameters ensure reward cycles are meaningful and provide miners with sufficient time to demonstrate contribution, while still allowing for responsive, dynamic reward updates.
TAO Hash (Subnet-14) is developed and maintained by Latent Holdings. Latent Holdings was founded in 2024 with a mission to accelerate the digital transformation of intelligence. Key team members include:
Joseph Jacks (JJ) – Co-Founder
JJ is the founder of OSS Capital, the first and only early-stage VC firm exclusively focused on commercial open source software (COSS) startups. Previously, he co-founded Kismatic, the first Kubernetes startup, and launched KubeCon, later donating it to the Linux Foundation as part of CNCF. Under his leadership, OSS Capital has backed 70+ companies supporting 150M+ users and 1M+ GitHub stars. JJ has led over 40 funding rounds, helping generate over $20B in value capture across sectors like AI, data, infrastructure, and developer tools.
Cameron Fairchild – Co-Founder
Cameron is a core contributor to the OpenTensor Foundation and is the co-founder of TAO Hash. He has a background in computer science (Univ. of Toronto) and previously worked on the OpenTensor Foundation. Profiles: GitHub camfairchild (Latent CTO); Twitter @KibibyteMe.
Benjamin Himes – Senior Engineer
Benjamin joins Laten Holdings from the Opentensor Foundation, the non-profit supporting the development of the Bittensor blockchain. At OTF, he played a key role in enhancing the Bittensor developer toolchain, including the SDK, CLI, and major upgrades like RAO and dTAO. His work focused on improving the developer experience and enabling scalable contributions to decentralized AI. Benjamin now continues this mission as part of a growing team dedicated to advancing the state of the art in artificial intelligence and open-source infrastructure.
Roman Chkhaidze – Engineer
A seasoned software developer with over 10 years of experience, specializing in Python and full-stack web development. Proficient in building robust applications using frameworks like Flask, FastAPI, Django, and Vue.js, with strong command of HTML5, CSS3, and modern JavaScript (ES6/TypeScript). Experienced across databases (PostgreSQL, MySQL, MongoDB, GCP), cloud platforms (AWS, GCE), and containerization tools (Docker, VMWare). Skilled in test automation with PyTest, Selenium, and more. Known for strong problem-solving, rapid tech adaptation, and driving process improvements across teams and environments.
Ibraheem Nadeem – Engineer
Michael Trestman – Technical Documentation lead.
Clément Blaise – Infrastructure
Xavier Lyu – Research
Yasmine Ibrahim – Compliance
Maciej Kula – Education



