🏆 CAP (Cardano Analytics Platform) has earned a Cardano Pie Award! #pieceofpie #hackathon Over the past 12 weeks, we built in public, shared our progress, and advanced the Cardano Analytics Platform through the Gimbalabs Piece of Pie programme. Thank you to @gimbalabs and @Cardano_CF for supporting builders turning ideas into real products. This is only the beginning. 🚀 #Cardano #Web3 #AI #BuildInPublic 🚀 Early access: cap.mobr.ai 📋 Closed Beta: cap.mobr.ai/beta 📂 Project presentation & demo videos: drive.google.com/drive/u/0/f…

Hello #Cardano family 👋 Week 12 (Final) of building CAP (Cardano Analytics Platform) at @gimbalabs #pieceofpie #hackathon As the Piece of Pie Hackathon comes to an end, we'd like to share what we've accomplished over the past 12 weeks and where CAP is headed next. 🚀 Early access: cap.mobr.ai 📋 Closed Beta: cap.mobr.ai/beta 📂 Project presentation & demo videos: drive.google.com/drive/u/0/f… What we built: AI & Analytics - Migrated CAP to a multi-agent AI architecture - Built federated analytics combining on-chain and off-chain data - Added natural language analytics powered by semantic reasoning - Developed a modular visualization engine for charts and analytical tables - Improved reliability, workflow orchestration, and infrastructure resilience Platform & Product - Implemented Cardano wallet authentication - Built payment-gated access and billing infrastructure - Added balance, entitlement, and usage management - Created operational admin tooling and monitoring capabilities - Improved deployment readiness and production architecture Community & User Experience - Launched our bot "BIL", CAP's Telegram interface - Introduced customizable workspace themes - Built and launched the Closed Beta program - Opened registrations to the Cardano community - Collected feedback from our first external users - Continued polishing the overall platform experience Why this matters: Our goal throughout the hackathon has been to transform CAP from a research prototype into a production-ready AI analytics platform for Cardano. Today, CAP combines agentic AI, federated on-chain/off-chain analytics, blockchain-native authentication and payments, rich visualizations, and conversational interfaces into a single platform designed to make Cardano data accessible to everyone. What's next: - Continue expanding the Closed Beta - Incorporate community feedback into the product roadmap - Further improve AI analytics and federated querying - Expand data sources and visualization capabilities - Continue building toward a production launch A huge thank you to @gimbalabs, the mentors, organizers, and the entire Cardano community for creating Piece of Pie and supporting builders throughout this journey. This may be the final hackathon update, but it's only the beginning for CAP. Let's keep building 🚀 No SQL. No setup. Just answers.

Hello #Cardano family 👋 Week 11 of building CAP (Cardano Analytics Platform) at @gimbalabs #pieceofpie #hackathon This week we focused on preparing CAP's Closed Beta launch by improving onboarding, streamlining authentication, and engaging the Cardano community as we begin welcoming our first external testers. 🚀 Apply for the Closed Beta (until July 18th): cap.mobr.ai/beta Early access: cap.mobr.ai What we did: Closed Beta program: * Launched the CAP Closed Beta campaign * Opened registrations for the first community testers * Shared the beta across the Cardano Forum, Cardano * Community Discord, and CAP Telegram channel * Improved the invitation and onboarding experience for beta participants * Added automatic approval for invited users during authentication * Refined multilingual beta invitation and confirmation email flows * Continued polishing the beta registration and operational workflows Platform improvements: * Simplified the authentication flow for invited beta users * Improved backend support for beta program operations * Refined off-chain metadata and AI context generation * Continued cleaning up the platform architecture and removing legacy components Frontend & UX: * Simplified the Cardano wallet selection experience * Removed unsupported and legacy wallet integrations * Updated the wallet interface to better reflect today's Cardano ecosystem * Continued usability improvements and frontend polish Why this matters: After several weeks of strengthening CAP's AI architecture, federated analytics engine, billing system, and operational tooling, we're now putting the platform in front of real users. The Closed Beta is our opportunity to validate the user experience, collect feedback from the Cardano community, and continue shaping CAP into a production-ready AI analytics platform. Next week: * Continue onboarding Closed Beta participants * Collect feedback from early testers * Refine the onboarding and analytics experience * Continue improving federated analytics workflows * Continue testing, optimization, and bug fixing Still building 🚀 No SQL. No setup. Just answers.

Hello #Cardano family 👋 Week 10 of building CAP (Cardano Analytics Platform) at @gimbalabs #pieceofpie #hackathon This week we focused on expanding CAP beyond on-chain analytics by introducing market data support, federated queries, and launching BIL — CAP's Telegram interface. What we did: BIL (Blockchain Intelligence Layer): * Launched BIL, CAP's Telegram bot interface * Enabled users to access CAP directly from Telegram * Added support for both one-to-one conversations and group/channel interactions * Continued improving the Telegram user experience and operational workflows Federated analytics: * Completed the first version of market data integration * Implemented federated query capabilities across multiple data sources * Enabled combined on-chain and market analytics workflows * Added support for analytics involving token activity, prices, and market indicators Platform improvements: * Continued hardening the analytics and query infrastructure * Improved the reliability of federated query execution * Continued testing and refining the end-to-end analytics experience * Fixed bugs and addressed edge cases discovered during development Why this matters: CAP is evolving from an on-chain analytics platform into a federated analytics engine capable of answering questions across blockchain and market data sources. With the launch of BIL, users can now access AI-powered Cardano analytics directly from Telegram, while federated queries open the door to richer insights that combine on-chain activity with market intelligence. Although CAP and BIL are still evolving, we're reaching an important milestone where analytics become accessible from anywhere, with no setup required. Next week: * Begin onboarding the first beta users * Gather feedback on the Telegram experience * Continue improving federated analytics capabilities * Refine the AI analytics workflows based on real-world usage * Continue testing, optimization, and bug fixing Still building 🚀 No SQL. No setup. Just answers. cap.mobr.ai

Hello #Cardano family 👋 Week 9 of building CAP (Cardano Analytics Platform) at @gimbalabs #pieceofpie #hackathon This week we focused on preparing CAP for its upcoming Closed Beta, while continuing to strengthen the platform's billing infrastructure and operational workflows. 🚀 Apply for the Closed Beta: cap.mobr.ai/beta What we did: Closed Beta program: * Built the complete Closed Beta registration workflow * Added the public Closed Beta landing page and onboarding flow * Implemented invitation and confirmation email workflows * Added an admin dashboard for managing beta registrations * Added admin alerts and notifications for new beta registrations * Added beta-user query notifications to better support early adopters * Introduced operational tooling to manage the growing beta community Platform & billing: * Improved Cardano payment reconciliation * Made payment reconciliation idempotent for greater reliability * Improved billing refresh after successful payment reconciliation * Continued hardening billing, notification, and operational workflows Frontend & UX: * Added a dedicated Closed Beta experience on the landing page * Improved admin interfaces for beta program management * Added a new CAP agent avatar to reinforce the platform identity * Continued UI polish and quality-of-life improvements Why this matters: After several weeks of strengthening CAP's AI architecture, federated analytics engine, visualization pipeline, and payment infrastructure, we're now preparing to onboard our first group of external users. The Closed Beta marks an important milestone as CAP evolves from an internal development platform into a production-ready AI-powered analytics service for the Cardano ecosystem. Next week: * Begin onboarding Closed Beta participants * Gather real-world feedback on analytics workflows * Continue improving billing and operational tooling * Refine the AI analytics experience based on early user feedback * Continue testing, optimization, and bug fixing Still building 🚀 No SQL. No setup. Just answers.

Hello #Cardano family 👋 Week 8 of building CAP (Cardano Analytics Platform) at @gimbalabs #pieceofpie #hackathon This week we focused on strengthening CAP's federated analytics engine, improving visualization quality, and further maturing the platform's architecture. Early access at: cap.mobr.ai What we did: Backend & query layer: * Expanded the federated analytics architecture for on-chain and off-chain data integration * Continued refining federated SPARQL + SQL query orchestration * Refactored SQL and SPARQL processing into dedicated federated service modules * Improved handling of infrastructure limits, timeouts, and transient failures * Added stronger differentiation between "no data" responses and infrastructure-related issues * Improved workflow status reporting and query execution feedback * Refined agentic workflow orchestration and status propagation * Expanded automated test coverage for federated analytics services * Continued improving reliability, maintainability, and separation of concerns across the backend Analytics & visualization: * Performed a major refactor of CAP's visualization architecture * Introduced dedicated visualization services and conversion layers * Added a Vega facade and modular chart-processing components * Improved separation of concerns across the visualization pipeline * Enhanced time-series and temporal data visualization handling * Improved heatmap rendering and axis labeling * Added richer metadata support for tables and chart outputs * Added date-aware formatting for analytical tables * Improved chart type detection and visualization consistency * Expanded support for textual analytical outputs alongside chart-based results Why this matters: CAP's mission is not only to answer questions about Cardano data, but to present insights in a form that is understandable, actionable, and visually meaningful. This week's work significantly improves how CAP processes, combines, and visualizes federated on-chain and off-chain data, while making the underlying analytics engine more robust, maintainable, and resilient to infrastructure constraints. Next week: * Continue improving federated analytics quality * Expand visualization capabilities and chart support * Refine agentic workflow reliability * Improve analytics presentation and user interpretation * Continue testing, optimization, and bug fixing Still building 🚀 No SQL. No setup. Just answers.

Hello #Cardano family 👋 Week 7 of building CAP (Cardano Analytics Platform) at @gimbalabs #pieceofpie #hackathon This week we focused on improving CAP’s operational tooling, federated analytics capabilities, and AI workflow reliability. cap.mobr.ai What we did: Backend & query layer: * Expanded federated SPARQL + SQL analytics capabilities * Added a federated result-processing layer for unified visualizations * Improved time-series and multi-series chart generation * Improved SQL/SPARQL result normalization for analytics workflows * Introduced a dedicated prompt-building layer to improve separation of concerns * Added automatic retries for transient LLM provider failures * Improved streaming response handling within the agentic workflow * Added infrastructure-aware handling for SPARQL rate limits and timeouts * Improved cache ingestion, pre-cache workflows, and similarity testing * Continued refining agentic orchestration, logging, and observability Admin & platform operations: * Added protection for admin-only data operation endpoints * Hardened backend controls around cache and raw data operations * Improved operational tooling for platform administration Frontend & UX: * Added a new Admin Data Console * Added interfaces for raw SPARQL execution, graph operations, and cache administration * Added dedicated Billing & Access pages * Improved separation between account settings and subscription management * Improved admin console feedback and usability * Refined navigation and sidebar layouts * Continued UI polish and bug fixing Why this matters: CAP is evolving into a complete Cardano intelligence platform, combining agentic AI, federated on-chain/off-chain analytics, production-grade administration tools, and blockchain-native access control. These improvements make the platform more reliable for end users while providing the operational capabilities needed to support a growing analytics service. Next week: * Continue improving federated analytics and visualization quality * Expand administrative tooling and observability * Improve billing and entitlement management * Refine agentic workflow reliability * Continue testing, optimization, and bug fixing Still building 🚀 No SQL. No setup. Just answers.

Hello #Cardano family ☀️ Week 6 of building CAP (Cardano Analytics Platform) at @gimbalabs #pieceofpie #hackathon Live on cap.mobr.ai This week we focused on improving CAP’s federated analytics reliability, refining Cardano-specific query handling, and continuing the frontend transition toward a more scalable, modular, theme-driven UI architecture. What we did: Backend & query layer: - Improved federated query handling and validation flows - Fixed edge cases where SQL results exist without SPARQL results - Improved query normalization and value extraction reliability - Improved Lovelace and ADA conversion handling in query results - Improved Cardano value detection and numeric interpretation - Improved entity-to-link conversion for Cardano explorer integrations - Improved Cardanoscan block-link mapping behavior - Fixed metrics persistence issues involving chart data serialization - Continued refining federated query naming consistency across agents, metrics, cache, planner, and LLM layers - Improved logging, typing consistency, cache handling, and internal agentic query tooling Frontend & visualization: - Continued migrating the CAP styling toward a token-driven theme architecture - Migrated dashboard, admin, analyses, shell, settings, loading, payment, and share modal styles to theme tokens - Modularized large CSS systems across navbar, sidebar, settings, analyses, and theme layers - Split Paper, Graphite, Midnight, and Terminal theme overrides into dedicated modules - Added theme auditing helpers for CSS size and hardcoded theme detection - Improved maintainability and separation of concerns across the frontend styling system - Added social-share thumbnail metadata for improved CAP link previews across platforms Why this matters: CAP is evolving into a more reliable and production-oriented AI analytics platform for Cardano, combining federated onchain/offchain querying, agentic orchestration, semantic reasoning, payment-enabled access control, and a customizable analytics workspace. This week’s work focused heavily on stability, maintainability, frontend scalability, and improving how CAP interprets and presents Cardano-native data. Next week: - Continue refining federated query orchestration - Improve semantic extraction and normalization reliability - Continue frontend modularization and token migration - Improve explorer/entity linking behavior - Continue testing billing and gated-access flows - Continue optimization, polish, and bug fixing Still building 🚀 No SQL. No setup. Just answers.

CAP just applied to CAP. 🤯 Beware @Cardano_CF, this may cause acronym instability in the @Cardano ecosystem. 😄 Yes. We just applied Cardano Analytics Platform to the Cardano Accelerator Program Fall ’26 Cohort, focused on Real-World Trust: verifying origins and data on Cardano. Perfectly named. Perfectly aligned. 😎 We’re building trusted AI analytics grounded in verifiable Cardano data. 🧠 🔗 cap.mobr.ai

Hello #Cardano family ☀️ Week 5 of building CAP (Cardano Analytics Platform) at @gimbalabs #pieceofpie #hackathon This week we focused on consolidating the progress from the last development cycle and pushing CAP further toward a production-ready Cardano intelligence platform, with stronger AI orchestration, payment-enabled access, cleaner Cardano infrastructure, and a more polished frontend experience. What we did: Backend & query layer: * Migrated the NL query pipeline toward a multi-agent LangGraph architecture * Added federated SPARQL + SQL query support for combined on-chain and OHLCV analytics * Introduced Cardano payment gating and billing infrastructure * Added DB-backed billing email notification settings and admin controls * Improved Cardano verification flow using PyCardano * Refactored Cardano integrations into a dedicated Cardano module for cleaner separation of concerns * Improved ontology modeling and internal query pipeline consistency * Continued backend cleanup, dependency hygiene, and production configuration improvements Frontend & visualization: * Added billing management and payment gating UI flows * Introduced user-selectable themes across the CAP interface * Added Terminal, Paper, Graphite, and Midnight visual variants * Improved theme coverage across authenticated workspace views * Improved theme isolation between public pages and authenticated workspace routes * Improved deployed sync-status parsing and frontend state consistency * Continued modularization and polish of Welcome, Admin, and workspace styling systems Admin & system UX: * Added admin controls for billing email notification management * Improved query details modal scrolling and overflow behavior * Unified modal scrollbar handling across admin interfaces * Improved NL query visualization behavior inside admin panels * Added safer streaming isolation to prevent cross-conversation message leakage * Improved browser storage detection and warning flows for restricted environments * Continued testing and bug fixing around billing, access, and deployment behavior Why this matters: CAP is evolving beyond a traditional RAG-style analytics interface into an agentic AI platform for Cardano data, combining semantic querying, federated on-chain/off-chain analytics, on-chain verification, billing infrastructure, admin controls, and customizable user experiences. The goal is to make Cardano analytics easier to access: no SQL setup, no infrastructure setup, just grounded answers from structured blockchain data. Next week: * Continue improving CAP’s semantic layer implementation * Keep refining federated query orchestration * Polish Billing & Access UX flows * Improve entitlement and query enforcement * Continue testing payment and balance-based usage flows * Expand sync-status observability * Improve production configuration checks * Continue frontend polish, optimization, and bug fixing Still building 🚀 No SQL. No setup. Just answers.

Hello #Cardano family ☀️ Week 4 of building CAP (Cardano Analytics Platform) at @gimbalabs #pieceofpie #hackathon This week we focused on transforming CAP into a production-grade agentic AI analytics platform and implementing the first working version of the platform access/payment layer. What we did: - Migrated CAP’s NL pipeline to a multi-agent LangGraph architecture - Replaced procedural orchestration with a stateful agent workflow - Introduced specialized agents for: * cache retrieval * planning * execution * critic/retry validation * contextualization * answer generation * persistence - Integrated: * LangGraph * LangChain * Semantic Kernel - Added self-healing retry flows through CriticAgent regeneration - Improved orchestration, execution reliability, and separation of concerns - Continued improving CAP’s semantic layer foundations On the payment/access side: - Implemented the first working version of CAP’s Cardano payment gateway foundation - Added Cardano wallet signature login flows - Implemented Premium access payment flows - Added ADA Balance deposit support - Added Premium activation from existing Balance - Added backend-enforced free query limits - Implemented quota consumption only after successful answers - Added billing/access status indicators in the UI - Improved paywall UX and frontend usability - Continued frontend polish and bug fixing Why this matters: CAP is evolving beyond a traditional RAG system into an agentic AI platform capable of orchestrating multiple reasoning, execution, and data-access layers across both onchain and offchain Cardano ecosystems. Next week: - Improve CAP’s semantic layer implementation - Continue refining federated query orchestration - Polish Billing & Access UX flows - Generalize currency/network handling for testnet/mainnet and future payment options - Connect free-limit paywalls directly to Premium/Balance upgrade actions - Improve add-balance and premium activation flows - Prepare explicit opt-in controls for auto-renewal and future pay-as-you-go usage - Continue testing, optimization, and bug fixing Still building 🚀 No SQL. No setup. Just answers.