With the amount of new subnets being added it can be hard to get up to date information across all subnets, so data may be slightly out of date from time to time
With the amount of new subnets being added it can be hard to get up to date information across all subnets, so data may be slightly out of date from time to time
Bitrecs (Bittensor Subnet 122) is a decentralized AI-powered recommendation engine focused on e-commerce. In simple terms, it provides online store owners – especially small and medium merchants – with an AI-driven “smart recommendations” system similar to what big retailers use (e.g. the “Customers also bought” suggestions on Amazon). The goal is to automatically show shoppers personalized product suggestions that boost key sales metrics like conversion rates, average order value (AOV), and customer lifetime value. By integrating Bitrecs into a shop, even a small merchant can leverage advanced AI to “unlock hidden sales” opportunities through tailored product recommendations. This means that instead of customers seeing only the items they explicitly search or click, the store will dynamically highlight other relevant products they are likely to buy, thereby increasing engagement and revenue. Bitrecs essentially levels the playing field by giving any merchant access to AI personalization usually reserved for retail giants, helping “merchants of all sizes” deliver real-time suggestions to their shoppers (which in turn drives more sales).
Once Bitrecs is installed, it continuously analyzes the store’s product catalog (and optionally past purchase data) to understand products and shopping patterns. When a customer is browsing the store (for example, viewing a product page or their cart), Bitrecs’ AI will generate recommendations for other items the customer might be interested in. These could be related products, frequently paired items, or personalized picks based on behavioral patterns. The recommendations update in real-time as the customer shops, and they are displayed directly on the storefront (e.g. a “Smart Product Recommendations” section). The aim is to expose more of the catalog to each shopper – if a visitor is fixated on one item or category, Bitrecs will suggest additional relevant products to encourage them to discover more and buy more. According to Bitrecs, this not only increases immediate sales but also improves long-term customer value by enhancing the shopping experience. Store owners don’t need to manually configure rules for what to recommend – the AI figures out meaningful product associations and customer preferences automatically. In summary, Bitrecs turns raw store data (catalog, inventory, pricing, etc.) into tailored product suggestions, helping merchants capture upsell and cross-sell opportunities that might otherwise be missed.
Encourages Product Discovery: By showing shoppers items related to what they’re viewing (or what’s in their cart), Bitrecs helps customers discover more of the store’s product range instead of focusing on a single item. This can surface complementary products or alternatives they might not have found on their own.
Increases Average Order Value (AOV): The AI often recommends items that pair well or add value, enticing shoppers to add extra products to their purchase. These personalized upsells and cross-sells lead to larger basket sizes on average. For example, someone buying running shoes might be shown high-quality socks or a fitness tracker – boosting the overall order value if added.
Boosts Conversions and Sales: By personalizing the shopping experience, Bitrecs makes it more likely that a visitor will find something they want to buy. The suggestions are aimed at being relevant and timely, which increases the chance of conversion (turning browsers into buyers). Stores using Bitrecs have reported higher conversion rates, more items sold per order, and greater revenue per customer as a result of these AI-driven recommendations.
Bitrecs (Bittensor Subnet 122) is a decentralized AI-powered recommendation engine focused on e-commerce. In simple terms, it provides online store owners – especially small and medium merchants – with an AI-driven “smart recommendations” system similar to what big retailers use (e.g. the “Customers also bought” suggestions on Amazon). The goal is to automatically show shoppers personalized product suggestions that boost key sales metrics like conversion rates, average order value (AOV), and customer lifetime value. By integrating Bitrecs into a shop, even a small merchant can leverage advanced AI to “unlock hidden sales” opportunities through tailored product recommendations. This means that instead of customers seeing only the items they explicitly search or click, the store will dynamically highlight other relevant products they are likely to buy, thereby increasing engagement and revenue. Bitrecs essentially levels the playing field by giving any merchant access to AI personalization usually reserved for retail giants, helping “merchants of all sizes” deliver real-time suggestions to their shoppers (which in turn drives more sales).
Once Bitrecs is installed, it continuously analyzes the store’s product catalog (and optionally past purchase data) to understand products and shopping patterns. When a customer is browsing the store (for example, viewing a product page or their cart), Bitrecs’ AI will generate recommendations for other items the customer might be interested in. These could be related products, frequently paired items, or personalized picks based on behavioral patterns. The recommendations update in real-time as the customer shops, and they are displayed directly on the storefront (e.g. a “Smart Product Recommendations” section). The aim is to expose more of the catalog to each shopper – if a visitor is fixated on one item or category, Bitrecs will suggest additional relevant products to encourage them to discover more and buy more. According to Bitrecs, this not only increases immediate sales but also improves long-term customer value by enhancing the shopping experience. Store owners don’t need to manually configure rules for what to recommend – the AI figures out meaningful product associations and customer preferences automatically. In summary, Bitrecs turns raw store data (catalog, inventory, pricing, etc.) into tailored product suggestions, helping merchants capture upsell and cross-sell opportunities that might otherwise be missed.
Encourages Product Discovery: By showing shoppers items related to what they’re viewing (or what’s in their cart), Bitrecs helps customers discover more of the store’s product range instead of focusing on a single item. This can surface complementary products or alternatives they might not have found on their own.
Increases Average Order Value (AOV): The AI often recommends items that pair well or add value, enticing shoppers to add extra products to their purchase. These personalized upsells and cross-sells lead to larger basket sizes on average. For example, someone buying running shoes might be shown high-quality socks or a fitness tracker – boosting the overall order value if added.
Boosts Conversions and Sales: By personalizing the shopping experience, Bitrecs makes it more likely that a visitor will find something they want to buy. The suggestions are aimed at being relevant and timely, which increases the chance of conversion (turning browsers into buyers). Stores using Bitrecs have reported higher conversion rates, more items sold per order, and greater revenue per customer as a result of these AI-driven recommendations.
In essence, what Bitrecs “does” is act as an intelligent sales assistant for online stores. It monitors what customers look at and leverages a network of AI models to decide which products to recommend next. This happens behind the scenes in a decentralized manner (more on that in the technical section), but to the merchant and shopper, it feels seamless – just like a built-in recommendation feature of the website. The benefit is a richer shopping experience for customers (they see more items they’ll likely love) and increased revenue for merchants (through higher conversion rates and bigger orders). The entire system is plug-and-play for store owners, requiring only a quick installation to start delivering AI-curated product suggestions.
Bitrecs isn’t just a single app – it’s a combination of a Shopify/WooCommerce plugin on the front-end and a decentralized AI network on the back-end. The “product” that a merchant interacts with is a plugin (or app) they install on their e-commerce platform, which then connects their store to the Bitrecs AI network. Under the hood, Bitrecs is implemented as a Bittensor subnet (number 122), meaning it runs on the Bittensor blockchain infrastructure and uses Bittensor’s decentralized machine learning framework for its recommendation engine. This architecture is what makes Bitrecs unique: unlike traditional recommendation engines that run on a centralized server or solely within a SaaS product, Bitrecs leverages a distributed network of AI “miners” and “validators” (nodes in the Bittensor network) to generate and vet recommendations.
In practical terms, here’s how the Bitrecs system works technically:
Plugin Installation & Data Sync: The merchant adds the Bitrecs plugin to their store (currently one-click install for Shopify, or a module for WooCommerce). Upon installation, the plugin connects to the Bitrecs network and shares key store data that the AI needs – mainly the product catalog (titles, descriptions, images, prices, stock) and, if the merchant opts in, some anonymized sales history for personalization. (No personal customer info is shared – Bitrecs only pulls product info and non-identifiable order patterns, per its privacy design.) This initial sync gives the AI a knowledge base of what products the store sells and some insight into which products are often bought together or by which season, etc.
AI Recommendation Generation (Decentralized Mining): When a shopper is on the site and a recommendation is needed (say viewing Product X on the store), the Bitrecs plugin sends a query to the Bittensor subnet asking for recommendations. This query can include context like the currently viewed product’s attributes, the user’s cart contents, or other behavioral signals. On the Bitrecs subnet, multiple AI “miner” nodes receive this query and each tries to generate a useful product recommendation in response. These miners are essentially independent AI models (or prompts to large language models) run by participants in the network. For example, one miner might be running a GPT-4-based model that was prompted with “Given a customer looking at product X (a running shoe), suggest another complementary product from the catalog.” Another miner might use a different model or strategy (some could use embeddings to find similar items, others might use collaborative filtering approaches with the provided data). The key is that each miner produces one or more suggested products that it believes the customer is likely to be interested in. Notably, miners can use various large language models (LLMs) and techniques – Bitrecs encourages miners to experiment with different LLMs and prompting methods to find the best recommendations, creating a competitive innovation environment.
Validation & Selection: Since many miners may return different suggestions, Bitrecs relies on a mechanism to evaluate which recommendation is best. This is where validator nodes come into play. In the Bitrecs subnet, validators receive the miner-generated recommendations and evaluate each response on quality and relevance. The evaluation criteria can include: Does the suggested product make sense given the query? Is it likely to appeal to the shopper? Is it diverse enough (not the same as what others suggested)? How fast was it generated (latency)? etc. The validators then rank or score the recommendations from all miners. Thanks to Bittensor’s consensus (the Yuma incentive mechanism), these scores will determine which miners get rewarded – but importantly, the top-ranked recommendation is selected to send back to the store. In essence, the decentralized network “agrees” on the best suggestion through this weighted voting process. This all happens in split-seconds. The Bitrecs subnet is optimized for real-time serving, so the customer isn’t kept waiting long for the recommendation to appear. Once the best recommendation is chosen, it’s returned via the plugin to the website.
Display to Shopper & Learning Feedback: The customer sees the Bitrecs-powered recommendation on the site (e.g. “You might also like: [Recommended Product]”). If they click it or add it to cart, that is a strong signal of a successful recommendation. Bitrecs can use such implicit feedback to improve over time. On the blockchain side, miners whose suggestions are frequently selected (and lead to positive outcomes) will accumulate higher weight, earning more dTAO rewards (Bitrecs operates with Bittensor’s token incentives) and effectively get reinforcement to continue with that strategy. Meanwhile, the Bitrecs team can also incorporate aggregated store data to refine the overall algorithms – for example, the plugin can report back conversion stats in aggregate. Bitrecs has a provision for continuous learning: it “continuously learns and improves over time” as more data flows through. The optional use of order history is part of this – if enabled, the system can learn from past purchase combinations to better suggest “customers who bought X also bought Y” patterns, all while respecting privacy (the data is anonymized and no personal customer info is retained). Over time, the network might even fine-tune or deploy specialized AI models that are discovered to work best for retail recommendations. (One stated goal of Bitrecs is to “discover the best and most consistent LLMs” for producing quality product recommendations, indicating that the network will evolve its AI models based on what proves most effective.) The decentralized nature ensures no single entity’s algorithm is the bottleneck – any miner in the world can try a new model or strategy, and if it produces better suggestions, the system will naturally start favoring that through the validator scoring and reward mechanism.
From a technical architecture perspective, Bitrecs is built atop the Opentensor/Bittensor framework, inheriting its robust design for scalable AI. Each Bitrecs “miner” node runs the Bitrecs subnet code (open-sourced on GitHub) which defines the prompting logic and how to interface with the store data. The subnet owner (the Bitrecs team) defined the incentive mechanism such that validators issue tasks (e.g. “Given context X, suggest a product”) and grade responses, and the blockchain’s consensus (Subtensor’s Yuma) handles reward distribution in $TAO or the subnet’s own token to reward good recommendations. The use of Bittensor’s blockchain means all the ranking and reward events are recorded on-chain, providing transparency. It’s worth noting that Bitrecs’s AI logic can involve multiple models: miners are free to use any large language model or algorithm that can parse the input text and output a product suggestion. Some miners might call external APIs or open-source models for NLP understanding of product descriptions. Others might eventually deploy fine-tuned models specifically trained on e-commerce recommendation tasks. The architecture is model-agnostic – it’s an open marketplace where the “best idea wins”, as determined by the collective validator judgement (which in turn is incentivized to align with actual recommendation quality). This is a fundamentally different approach from a closed, single-model system. It means Bitrecs’ recommendation engine is continuously crowdsourced and optimized by a decentralized community of AI developers, all competing and cooperating to serve the most relevant product suggestions. For the store owner, however, all of this complexity is abstracted away: they simply see an app that plugs into their site and starts giving AI-driven recommendations. The heavy lifting (querying models, ranking outputs, learning from results) is handled by the Bitrecs subnet on the back-end.
In addition to the core recommendation engine, Bitrecs provides a dashboard and analytics for merchants (via the plugin interface). Store owners can see metrics like how many recommendations were clicked, which products are most often recommended, and the lift in conversion/AOV attributed to Bitrecs. The integration is designed to be as frictionless as possible – on Shopify it’s a one-click app install with automatic theme integration, and on WooCommerce it’s a simple plugin download. Bitrecs also allows some customization (for instance, merchants can choose where on their site the recommendations widget appears and select from different recommendation “modes” or layouts). The pricing model, as seen on the Bitrecs website, currently offers a Free tier (for small stores) and a Pro tier for larger stores, which differ by how many products and orders can be handled per month. This implies that the network and the team have accounted for scaling – larger stores with thousands of products or unlimited requests would require the Pro plan, possibly to cover the computational costs on the network. Technically, this also suggests the Bitrecs subnet can scale its throughput based on demand (more miners can join if there’s lots of request volume, etc.).
To summarize the build: Bitrecs = [Shopify/WooCommerce front-end app] + [Bittensor Subnet back-end]. The front-end captures store context and displays results; the back-end is where decentralized AI inference and ranking happens. This product is one of the first real-world use-cases of Bittensor in e-commerce — it effectively bridges Web2 commerce platforms with a Web3 AI network. The architecture ensures that no centralized company is solely dictating the recommendations; instead, the intelligence comes from a network that any developer can contribute to (by running a node). Yet, the Bitrecs team oversees the subnet (maintaining the code, setting the rules of the game, and integrating it with commerce platforms). This blend of AI-as-a-service and decentralized infrastructure is what makes Bitrecs a novel build in the AI commerce space.
In essence, what Bitrecs “does” is act as an intelligent sales assistant for online stores. It monitors what customers look at and leverages a network of AI models to decide which products to recommend next. This happens behind the scenes in a decentralized manner (more on that in the technical section), but to the merchant and shopper, it feels seamless – just like a built-in recommendation feature of the website. The benefit is a richer shopping experience for customers (they see more items they’ll likely love) and increased revenue for merchants (through higher conversion rates and bigger orders). The entire system is plug-and-play for store owners, requiring only a quick installation to start delivering AI-curated product suggestions.
Bitrecs isn’t just a single app – it’s a combination of a Shopify/WooCommerce plugin on the front-end and a decentralized AI network on the back-end. The “product” that a merchant interacts with is a plugin (or app) they install on their e-commerce platform, which then connects their store to the Bitrecs AI network. Under the hood, Bitrecs is implemented as a Bittensor subnet (number 122), meaning it runs on the Bittensor blockchain infrastructure and uses Bittensor’s decentralized machine learning framework for its recommendation engine. This architecture is what makes Bitrecs unique: unlike traditional recommendation engines that run on a centralized server or solely within a SaaS product, Bitrecs leverages a distributed network of AI “miners” and “validators” (nodes in the Bittensor network) to generate and vet recommendations.
In practical terms, here’s how the Bitrecs system works technically:
Plugin Installation & Data Sync: The merchant adds the Bitrecs plugin to their store (currently one-click install for Shopify, or a module for WooCommerce). Upon installation, the plugin connects to the Bitrecs network and shares key store data that the AI needs – mainly the product catalog (titles, descriptions, images, prices, stock) and, if the merchant opts in, some anonymized sales history for personalization. (No personal customer info is shared – Bitrecs only pulls product info and non-identifiable order patterns, per its privacy design.) This initial sync gives the AI a knowledge base of what products the store sells and some insight into which products are often bought together or by which season, etc.
AI Recommendation Generation (Decentralized Mining): When a shopper is on the site and a recommendation is needed (say viewing Product X on the store), the Bitrecs plugin sends a query to the Bittensor subnet asking for recommendations. This query can include context like the currently viewed product’s attributes, the user’s cart contents, or other behavioral signals. On the Bitrecs subnet, multiple AI “miner” nodes receive this query and each tries to generate a useful product recommendation in response. These miners are essentially independent AI models (or prompts to large language models) run by participants in the network. For example, one miner might be running a GPT-4-based model that was prompted with “Given a customer looking at product X (a running shoe), suggest another complementary product from the catalog.” Another miner might use a different model or strategy (some could use embeddings to find similar items, others might use collaborative filtering approaches with the provided data). The key is that each miner produces one or more suggested products that it believes the customer is likely to be interested in. Notably, miners can use various large language models (LLMs) and techniques – Bitrecs encourages miners to experiment with different LLMs and prompting methods to find the best recommendations, creating a competitive innovation environment.
Validation & Selection: Since many miners may return different suggestions, Bitrecs relies on a mechanism to evaluate which recommendation is best. This is where validator nodes come into play. In the Bitrecs subnet, validators receive the miner-generated recommendations and evaluate each response on quality and relevance. The evaluation criteria can include: Does the suggested product make sense given the query? Is it likely to appeal to the shopper? Is it diverse enough (not the same as what others suggested)? How fast was it generated (latency)? etc. The validators then rank or score the recommendations from all miners. Thanks to Bittensor’s consensus (the Yuma incentive mechanism), these scores will determine which miners get rewarded – but importantly, the top-ranked recommendation is selected to send back to the store. In essence, the decentralized network “agrees” on the best suggestion through this weighted voting process. This all happens in split-seconds. The Bitrecs subnet is optimized for real-time serving, so the customer isn’t kept waiting long for the recommendation to appear. Once the best recommendation is chosen, it’s returned via the plugin to the website.
Display to Shopper & Learning Feedback: The customer sees the Bitrecs-powered recommendation on the site (e.g. “You might also like: [Recommended Product]”). If they click it or add it to cart, that is a strong signal of a successful recommendation. Bitrecs can use such implicit feedback to improve over time. On the blockchain side, miners whose suggestions are frequently selected (and lead to positive outcomes) will accumulate higher weight, earning more dTAO rewards (Bitrecs operates with Bittensor’s token incentives) and effectively get reinforcement to continue with that strategy. Meanwhile, the Bitrecs team can also incorporate aggregated store data to refine the overall algorithms – for example, the plugin can report back conversion stats in aggregate. Bitrecs has a provision for continuous learning: it “continuously learns and improves over time” as more data flows through. The optional use of order history is part of this – if enabled, the system can learn from past purchase combinations to better suggest “customers who bought X also bought Y” patterns, all while respecting privacy (the data is anonymized and no personal customer info is retained). Over time, the network might even fine-tune or deploy specialized AI models that are discovered to work best for retail recommendations. (One stated goal of Bitrecs is to “discover the best and most consistent LLMs” for producing quality product recommendations, indicating that the network will evolve its AI models based on what proves most effective.) The decentralized nature ensures no single entity’s algorithm is the bottleneck – any miner in the world can try a new model or strategy, and if it produces better suggestions, the system will naturally start favoring that through the validator scoring and reward mechanism.
From a technical architecture perspective, Bitrecs is built atop the Opentensor/Bittensor framework, inheriting its robust design for scalable AI. Each Bitrecs “miner” node runs the Bitrecs subnet code (open-sourced on GitHub) which defines the prompting logic and how to interface with the store data. The subnet owner (the Bitrecs team) defined the incentive mechanism such that validators issue tasks (e.g. “Given context X, suggest a product”) and grade responses, and the blockchain’s consensus (Subtensor’s Yuma) handles reward distribution in $TAO or the subnet’s own token to reward good recommendations. The use of Bittensor’s blockchain means all the ranking and reward events are recorded on-chain, providing transparency. It’s worth noting that Bitrecs’s AI logic can involve multiple models: miners are free to use any large language model or algorithm that can parse the input text and output a product suggestion. Some miners might call external APIs or open-source models for NLP understanding of product descriptions. Others might eventually deploy fine-tuned models specifically trained on e-commerce recommendation tasks. The architecture is model-agnostic – it’s an open marketplace where the “best idea wins”, as determined by the collective validator judgement (which in turn is incentivized to align with actual recommendation quality). This is a fundamentally different approach from a closed, single-model system. It means Bitrecs’ recommendation engine is continuously crowdsourced and optimized by a decentralized community of AI developers, all competing and cooperating to serve the most relevant product suggestions. For the store owner, however, all of this complexity is abstracted away: they simply see an app that plugs into their site and starts giving AI-driven recommendations. The heavy lifting (querying models, ranking outputs, learning from results) is handled by the Bitrecs subnet on the back-end.
In addition to the core recommendation engine, Bitrecs provides a dashboard and analytics for merchants (via the plugin interface). Store owners can see metrics like how many recommendations were clicked, which products are most often recommended, and the lift in conversion/AOV attributed to Bitrecs. The integration is designed to be as frictionless as possible – on Shopify it’s a one-click app install with automatic theme integration, and on WooCommerce it’s a simple plugin download. Bitrecs also allows some customization (for instance, merchants can choose where on their site the recommendations widget appears and select from different recommendation “modes” or layouts). The pricing model, as seen on the Bitrecs website, currently offers a Free tier (for small stores) and a Pro tier for larger stores, which differ by how many products and orders can be handled per month. This implies that the network and the team have accounted for scaling – larger stores with thousands of products or unlimited requests would require the Pro plan, possibly to cover the computational costs on the network. Technically, this also suggests the Bitrecs subnet can scale its throughput based on demand (more miners can join if there’s lots of request volume, etc.).
To summarize the build: Bitrecs = [Shopify/WooCommerce front-end app] + [Bittensor Subnet back-end]. The front-end captures store context and displays results; the back-end is where decentralized AI inference and ranking happens. This product is one of the first real-world use-cases of Bittensor in e-commerce — it effectively bridges Web2 commerce platforms with a Web3 AI network. The architecture ensures that no centralized company is solely dictating the recommendations; instead, the intelligence comes from a network that any developer can contribute to (by running a node). Yet, the Bitrecs team oversees the subnet (maintaining the code, setting the rules of the game, and integrating it with commerce platforms). This blend of AI-as-a-service and decentralized infrastructure is what makes Bitrecs a novel build in the AI commerce space.
The Bitrecs project appears to be driven by a small, highly-focused team within the Bittensor community. While detailed team identities have not been very public, we do know that Bitrecs was founded in 2024 as one of Bittensor’s specialized subnets addressing a real-world vertical (e-commerce recommendations). The team behind Bitrecs is likely a mix of AI engineers and e-commerce tech experts. Their backgrounds combine machine learning (particularly NLP/Large Language Models) and blockchain/Web3 development, which aligns with the skill set needed to build a Bittensor subnet and a Shopify app together. There are indications that the Bitrecs founders/contributors have been involved in the broader crypto AI space – for instance, the project’s social media is followed by known Bittensor community members. The Bitrecs Twitter (X) account is relatively new (with a modest follower count, suggesting the team is still growing its presence) and it mainly shares updates about the subnet’s status and features.
As of mid-2025, specific individual names (like CEO or CTO of Bitrecs) have not been officially announced on the website or documentation available. The Subnet Owner in Bittensor terms is presumably the Bitrecs founding team or company, which would be receiving a portion of the subnet’s token emissions. This suggests there is a formal entity or founder overseeing the project’s development and business operations. Some community sleuthing has linked certain pseudonymous contributors on X (Twitter) to Bitrecs – for example, a user with interests in AI and DeFi who claims involvement in Bitrecs – but without official confirmation, it’s safe to say the team is intentionally keeping a low profile publicly, at least until the product gains traction. What we can infer from their output is that the team is technically adept (given they delivered a working Bittensor subnet and Shopify integration) and product-minded (they’ve polished a plugin with a user-friendly interface, pricing tiers, etc.). They likely have at least one AI scientist/engineer who crafted the recommendation logic and one full-stack or blockchain developer who handled the subnet coding and integration.
The team’s background likely includes prior projects in AI or crypto – for instance, involvement in other AI startups or contributions in the Bittensor ecosystem. While names aren’t public, the effectiveness of Bitrecs’ early product speaks to the team’s competence. We can expect that as Bitrecs grows, the team will become more visible, potentially revealing advisors or investors. To contact or learn about the team, currently one would go through official channels like the Bitrecs website or their X (Twitter) account. They have a support email and likely a presence on Bittensor’s community Discord under the Bitrecs subnet channel where the developers might interact with miners and users. Overall, even with limited public info, it’s clear the Bitrecs team is passionate about bridging AI and e-commerce and is leveraging Bittensor’s decentralized tech to do so. They stand as pioneers in applying decentralized AI to a mainstream use-case, which speaks to their vision and technical savvy. As the project matures, we anticipate more details on the individuals behind Bitrecs and their roadmap forward.
The Bitrecs project appears to be driven by a small, highly-focused team within the Bittensor community. While detailed team identities have not been very public, we do know that Bitrecs was founded in 2024 as one of Bittensor’s specialized subnets addressing a real-world vertical (e-commerce recommendations). The team behind Bitrecs is likely a mix of AI engineers and e-commerce tech experts. Their backgrounds combine machine learning (particularly NLP/Large Language Models) and blockchain/Web3 development, which aligns with the skill set needed to build a Bittensor subnet and a Shopify app together. There are indications that the Bitrecs founders/contributors have been involved in the broader crypto AI space – for instance, the project’s social media is followed by known Bittensor community members. The Bitrecs Twitter (X) account is relatively new (with a modest follower count, suggesting the team is still growing its presence) and it mainly shares updates about the subnet’s status and features.
As of mid-2025, specific individual names (like CEO or CTO of Bitrecs) have not been officially announced on the website or documentation available. The Subnet Owner in Bittensor terms is presumably the Bitrecs founding team or company, which would be receiving a portion of the subnet’s token emissions. This suggests there is a formal entity or founder overseeing the project’s development and business operations. Some community sleuthing has linked certain pseudonymous contributors on X (Twitter) to Bitrecs – for example, a user with interests in AI and DeFi who claims involvement in Bitrecs – but without official confirmation, it’s safe to say the team is intentionally keeping a low profile publicly, at least until the product gains traction. What we can infer from their output is that the team is technically adept (given they delivered a working Bittensor subnet and Shopify integration) and product-minded (they’ve polished a plugin with a user-friendly interface, pricing tiers, etc.). They likely have at least one AI scientist/engineer who crafted the recommendation logic and one full-stack or blockchain developer who handled the subnet coding and integration.
The team’s background likely includes prior projects in AI or crypto – for instance, involvement in other AI startups or contributions in the Bittensor ecosystem. While names aren’t public, the effectiveness of Bitrecs’ early product speaks to the team’s competence. We can expect that as Bitrecs grows, the team will become more visible, potentially revealing advisors or investors. To contact or learn about the team, currently one would go through official channels like the Bitrecs website or their X (Twitter) account. They have a support email and likely a presence on Bittensor’s community Discord under the Bitrecs subnet channel where the developers might interact with miners and users. Overall, even with limited public info, it’s clear the Bitrecs team is passionate about bridging AI and e-commerce and is leveraging Bittensor’s decentralized tech to do so. They stand as pioneers in applying decentralized AI to a mainstream use-case, which speaks to their vision and technical savvy. As the project matures, we anticipate more details on the individuals behind Bitrecs and their roadmap forward.
Bitrecs is a relatively new project, and its roadmap reflects an early-stage but ambitious trajectory. Initially conceptualized and developed in the first half of 2024, Bitrecs spent its early phase building the core product (the subnet code and Shopify integration) and testing it in a limited environment. By mid-2024, the concept was introduced to the community (CryptoZPunisher’s June 2024 post served as an announcement of Subnet 122: Bitrecs, noting that it was not yet live at that time). The first milestone on the roadmap was achieving a functional MVP (Minimum Viable Product) – essentially proving that a Bittensor subnet could successfully deliver real-time recommendations to a Shopify store.
Going into 2025, the roadmap focus is on expansion and refinement. One major goal is to onboard more e-commerce platforms: having already built for Shopify (which powers millions of online shops) and WooCommerce (another widely used platform), Bitrecs may look next at platforms like Magento, BigCommerce, or others. Supporting additional platforms would widen the user base. Each platform integration requires some development work (for example, a plugin or app specific to that ecosystem), so these would be sequential targets on the roadmap. We know WooCommerce integration is in progress (as a plugin download), which likely became available in 2025. A reasonable roadmap item is to officially publish the Shopify App on the Shopify App Store (currently, merchants might install it via a link or directly – getting listed in the App Store could significantly increase adoption). The Shopify App Store listing might require certain security and compliance reviews, so achieving “Built for Shopify” status could be a 2025 goal for the team.
On the AI side, the roadmap emphasizes continuous improvement of recommendation quality. The Bitrecs subnet is designed to “discover the best and most consistent LLMs” for recommendations – in practice, this means throughout 2025 the team will monitor which approaches miners take and what yields the highest conversion lifts. If certain open-source models perform exceptionally well, the team might collaborate with those miners or even incorporate those models more directly (e.g., providing a reference model for new miners to use). There’s also a potential plan for a custom fine-tuned model: Bitrecs could train its own recommendation-specific language model using aggregated data from many stores. This would be a longer-term project, but it could dramatically improve the quality of suggestions and reduce reliance on external APIs (like OpenAI) some miners might be using. The decentralized nature means the model could even be distributed for miners to run. We can expect research and possibly a paper or blog post on “AI techniques for decentralized recommendations” as Bitrecs reaches technical maturity – they are trailblazing in this niche, so documenting their methods might be on the roadmap to build credibility.
Another key roadmap item is enhancing the feedback loop and incentive mechanism. In a traditional recommender system, you continuously update your model based on which recommendations led to sales. In Bitrecs, this involves both on-chain mechanisms (validators weighting miners) and off-chain analytics. The team likely plans to incorporate conversion feedback into the subnet’s incentive model more directly. For instance, they could adjust validator scoring algorithms to factor in known purchase correlations or even implement a system where the plugin reports back when a recommended item was purchased, and that could translate into an on-chain reward signal. Such integration of real sales data into the blockchain rewards would be cutting-edge, essentially creating a closed-loop where the AI is directly rewarded for generating revenue. Achieving this would be a big milestone and might appear on the roadmap as “Phase 2: Integrate conversion or A/B test results into subnet incentives.”
From a business perspective, the roadmap likely includes growing merchant adoption. In 2025, Bitrecs might aim to get a certain number of stores or a certain GMV (Gross Merchandise Volume) using the system to validate its efficacy. This could involve marketing efforts in the Shopify ecosystem (perhaps appearing in Shopify’s app spotlight or partnerships with agencies). The mention of “Maximizing your marketing ROI” in Bitrecs materials suggests they are pitching this as a must-have tool for e-commerce marketing. So, part of the roadmap could be to gather case studies – for example, Show that Store ABC saw a +X% increase in sales after using Bitrecs. Success stories will help drive wider adoption. Thus, an objective might be: By end of 2025, onboard 100+ active stores and demonstrate an average lift in conversion/AOV by at least Y%.
One possible future feature is an AI shopping assistant or chatbot that uses the Bitrecs network to handle customer queries (e.g., “Which running shoes do you recommend for trail running?” and the assistant suggests products). This would leverage the same backend but in a conversational format. While not officially stated, it aligns with the general direction of AI in e-commerce. The team might explore this once the core recommendation widgets are well-established. Another feature could be email/push recommendations – Bitrecs could generate personalized product picks for email newsletters or retargeting campaigns, again using its AI knowledge. Essentially, expanding the touchpoints where Bitrecs recommendations appear could be on the roadmap (in-site, in-app, email, etc.).
In terms of timeline, the short-term (next 6 months) is likely about polishing the product and scaling to more stores. The medium-term (6-12 months) is about improving the AI models and possibly introducing new recommendation types (like bundle recommendations, or more granular personalization if customers log in). The long-term (12+ months) could involve solidifying Bitrecs as the go-to decentralized recommendation solution, possibly even moving into physical retail (an extreme case: imagine if point-of-sale systems or kiosk apps use Bitrecs to recommend items – that’s far-out but shows the extensibility). For now, a concrete long-term roadmap item is simply staying adaptive: Bitrecs will continuously incorporate the latest LLM advances. Given how fast AI is evolving, if a new model (say GPT-5 or an open-source equivalent) emerges that significantly improves recommendation quality, Bitrecs would aim to integrate that via its miners. The decentralized setup makes this relatively agile (community miners will likely adopt new models as they come, without the core team having to rewrite everything).
To conclude, the Bitrecs roadmap is about scaling up and optimizing: more platform integrations, more merchants, more data – and using that data to fine-tune the AI and incentives. The project has already moved from concept to deployment within about a year, which is a strong pace. The next milestones will be measured in adoption and performance improvements. Users can expect frequent updates from the team in the form of blog posts or social media threads highlighting progress (for example, “Bitrecs achieved +30% lift for merchants this quarter” or “Now supporting Magento stores!”). Being part of the Bittensor ecosystem, Bitrecs’ progress is often mentioned in community updates, and it contributes to the overall narrative of decentralized AI finding real use. If Bitrecs succeeds, its roadmap could even include spinning off into a full-fledged platform for various AI-driven e-commerce services (recommendations today, perhaps pricing optimization or inventory predictions tomorrow, as additional subnet functionalities). Those speculative ideas aside, the clear near-term roadmap is: grow the network of miners & merchants, iterate on the AI for better recommendations, and cement Bitrecs as a reliable, high-performing recommendation engine in the e-commerce world. Each step forward for Bitrecs not only enhances the product itself but also serves as a proof-of-concept for Bittensor’s vision of decentralized AI services tackling real business problems.
Bitrecs is a relatively new project, and its roadmap reflects an early-stage but ambitious trajectory. Initially conceptualized and developed in the first half of 2024, Bitrecs spent its early phase building the core product (the subnet code and Shopify integration) and testing it in a limited environment. By mid-2024, the concept was introduced to the community (CryptoZPunisher’s June 2024 post served as an announcement of Subnet 122: Bitrecs, noting that it was not yet live at that time). The first milestone on the roadmap was achieving a functional MVP (Minimum Viable Product) – essentially proving that a Bittensor subnet could successfully deliver real-time recommendations to a Shopify store.
Going into 2025, the roadmap focus is on expansion and refinement. One major goal is to onboard more e-commerce platforms: having already built for Shopify (which powers millions of online shops) and WooCommerce (another widely used platform), Bitrecs may look next at platforms like Magento, BigCommerce, or others. Supporting additional platforms would widen the user base. Each platform integration requires some development work (for example, a plugin or app specific to that ecosystem), so these would be sequential targets on the roadmap. We know WooCommerce integration is in progress (as a plugin download), which likely became available in 2025. A reasonable roadmap item is to officially publish the Shopify App on the Shopify App Store (currently, merchants might install it via a link or directly – getting listed in the App Store could significantly increase adoption). The Shopify App Store listing might require certain security and compliance reviews, so achieving “Built for Shopify” status could be a 2025 goal for the team.
On the AI side, the roadmap emphasizes continuous improvement of recommendation quality. The Bitrecs subnet is designed to “discover the best and most consistent LLMs” for recommendations – in practice, this means throughout 2025 the team will monitor which approaches miners take and what yields the highest conversion lifts. If certain open-source models perform exceptionally well, the team might collaborate with those miners or even incorporate those models more directly (e.g., providing a reference model for new miners to use). There’s also a potential plan for a custom fine-tuned model: Bitrecs could train its own recommendation-specific language model using aggregated data from many stores. This would be a longer-term project, but it could dramatically improve the quality of suggestions and reduce reliance on external APIs (like OpenAI) some miners might be using. The decentralized nature means the model could even be distributed for miners to run. We can expect research and possibly a paper or blog post on “AI techniques for decentralized recommendations” as Bitrecs reaches technical maturity – they are trailblazing in this niche, so documenting their methods might be on the roadmap to build credibility.
Another key roadmap item is enhancing the feedback loop and incentive mechanism. In a traditional recommender system, you continuously update your model based on which recommendations led to sales. In Bitrecs, this involves both on-chain mechanisms (validators weighting miners) and off-chain analytics. The team likely plans to incorporate conversion feedback into the subnet’s incentive model more directly. For instance, they could adjust validator scoring algorithms to factor in known purchase correlations or even implement a system where the plugin reports back when a recommended item was purchased, and that could translate into an on-chain reward signal. Such integration of real sales data into the blockchain rewards would be cutting-edge, essentially creating a closed-loop where the AI is directly rewarded for generating revenue. Achieving this would be a big milestone and might appear on the roadmap as “Phase 2: Integrate conversion or A/B test results into subnet incentives.”
From a business perspective, the roadmap likely includes growing merchant adoption. In 2025, Bitrecs might aim to get a certain number of stores or a certain GMV (Gross Merchandise Volume) using the system to validate its efficacy. This could involve marketing efforts in the Shopify ecosystem (perhaps appearing in Shopify’s app spotlight or partnerships with agencies). The mention of “Maximizing your marketing ROI” in Bitrecs materials suggests they are pitching this as a must-have tool for e-commerce marketing. So, part of the roadmap could be to gather case studies – for example, Show that Store ABC saw a +X% increase in sales after using Bitrecs. Success stories will help drive wider adoption. Thus, an objective might be: By end of 2025, onboard 100+ active stores and demonstrate an average lift in conversion/AOV by at least Y%.
One possible future feature is an AI shopping assistant or chatbot that uses the Bitrecs network to handle customer queries (e.g., “Which running shoes do you recommend for trail running?” and the assistant suggests products). This would leverage the same backend but in a conversational format. While not officially stated, it aligns with the general direction of AI in e-commerce. The team might explore this once the core recommendation widgets are well-established. Another feature could be email/push recommendations – Bitrecs could generate personalized product picks for email newsletters or retargeting campaigns, again using its AI knowledge. Essentially, expanding the touchpoints where Bitrecs recommendations appear could be on the roadmap (in-site, in-app, email, etc.).
In terms of timeline, the short-term (next 6 months) is likely about polishing the product and scaling to more stores. The medium-term (6-12 months) is about improving the AI models and possibly introducing new recommendation types (like bundle recommendations, or more granular personalization if customers log in). The long-term (12+ months) could involve solidifying Bitrecs as the go-to decentralized recommendation solution, possibly even moving into physical retail (an extreme case: imagine if point-of-sale systems or kiosk apps use Bitrecs to recommend items – that’s far-out but shows the extensibility). For now, a concrete long-term roadmap item is simply staying adaptive: Bitrecs will continuously incorporate the latest LLM advances. Given how fast AI is evolving, if a new model (say GPT-5 or an open-source equivalent) emerges that significantly improves recommendation quality, Bitrecs would aim to integrate that via its miners. The decentralized setup makes this relatively agile (community miners will likely adopt new models as they come, without the core team having to rewrite everything).
To conclude, the Bitrecs roadmap is about scaling up and optimizing: more platform integrations, more merchants, more data – and using that data to fine-tune the AI and incentives. The project has already moved from concept to deployment within about a year, which is a strong pace. The next milestones will be measured in adoption and performance improvements. Users can expect frequent updates from the team in the form of blog posts or social media threads highlighting progress (for example, “Bitrecs achieved +30% lift for merchants this quarter” or “Now supporting Magento stores!”). Being part of the Bittensor ecosystem, Bitrecs’ progress is often mentioned in community updates, and it contributes to the overall narrative of decentralized AI finding real use. If Bitrecs succeeds, its roadmap could even include spinning off into a full-fledged platform for various AI-driven e-commerce services (recommendations today, perhaps pricing optimization or inventory predictions tomorrow, as additional subnet functionalities). Those speculative ideas aside, the clear near-term roadmap is: grow the network of miners & merchants, iterate on the AI for better recommendations, and cement Bitrecs as a reliable, high-performing recommendation engine in the e-commerce world. Each step forward for Bitrecs not only enhances the product itself but also serves as a proof-of-concept for Bittensor’s vision of decentralized AI services tackling real business problems.
Bitrecs is a novel recommendation engine built on the Bittensor network
Our team’s extensive 30+ year cumulative background in ecommerce, swe and crypto (17’) uniquely positions us for success in Bittensor.
Once we push and validate v2 scoring we will turn to our existing network of stores and simultaneously turn the marketing machine on.
Bitrecs.
Hey do you know what this is?
With Bitrecs you'll never need to. Bring frontier AI models like chatGPT, Gemini and more to your ecommerce store in a few clicks.
122.
We’re building systems on the verge of what’s possible
Contribute your intelligence to the future of ecommerce:
GitHub - bitrecs/bitrecs-subnet: Bitrecs product recommendations
Bitrecs product recommendations. Contribute to bitrecs/bitrecs-subnet development by creating an account on GitHub.
github.com
The problem with AI isn't intelligence, it’s confidence.
When a single AI suggests a product, how confident should you be? When 10+ models agree on the same recommendation? That's a consensus you can trust.
Bitrecs does this automatically on every query, in real-time.
122.
E-commerce isn't waiting for halving cycles.
$6.1T flows through online stores every year. Bitrecs taps into that stream directly.
Bittensor rewards incentives.
E-commerce rewards results.
