Today was mostly about making Hack-A-Gent more reliable instead of adding new features. I finally tracked down a long-standing parser issue that turned out to be an architectural problem: the CLI and the analysis pipeline were using two completely different parsers. I merged everything into a single parser, updated the production build to include it, and verified that all parser regression tests and the full test suite still pass. After that, I confirmed the CLI now extracts real judging criteria like Impact and Innovation instead of mistakenly treating navigation links as judging criteria. With parsing in a much better place, I shifted focus toward hardening the overall pipeline by planning phased reliability work around generation, auto-repair, validation, and recovery so the goal is no longer just generating code, but consistently producing projects that typecheck, build, run, and are ready for submission.

This week I focused on making Hack-A-Gent stable and release-ready instead of adding new features. I cleaned up dead code, removed unused files, improved CLI logging and error messages, and verified the project with a full build and test pass. The CLI now parses real hackathon pages, plans projects, and gracefully falls back when AI providers are unavailable. I also ran end-to-end tests on real hackathon URLs to identify the final issues before release. The remaining work is improving template fallback and validation so every generated project is immediately usable, bringing Hack-A-Gent closer to its first stable release.

Might be on final stages, making sure the agent doesn’t blows up in users hands. Taking time but needs to be done, Hope I can ship it by this week

Spent today improving Hack-A-Gent’s long-term direction instead of just fixing bugs. I mapped out a roadmap to turn it into a real Devpost hackathon agent, including a hackathon qualification engine, real Devpost benchmark suite, autonomous repair loop, browser validation, and organizational learning. Also audited the current architecture to identify what already exists, what needs refactoring, and what should be built next before writing more code.

I built this project using my AI (Actual Intelligence) and a really well-written Stardance guide. It was my first time working with the NASA APOD API, and I learned a lot about fetch(), Vite, and environment variables. I also ran into a GitHub Pages issue that took some debugging before I finally got everything working. The CSS was surprisingly fun too, especially experimenting with clip-path and Google Fonts. Overall, this was a simple project that taught me a lot more than I expected.

The website generation is pretty irritation, partialy fixed it. Needs more work

Hack-A-Gent

I built Hack-A-Gent, a CLI-based AI agent that helps developers turn a Devpost hackathon challenge into a working project in minutes. Simply paste a Devpost competition URL into the terminal, and Hack-A-Gent analyzes the challenge, extracts the judging criteria and submission requirements, plans an appropriate architecture, and generates a production-ready project scaffold.

You bring your own LLM API key (Anthropic, OpenAI, Google Gemini, NVIDIA NIMs, Ollama, and more), and Hack-A-Gent orchestrates the rest—from understanding the hackathon prompt to generating code, documentation, and a submission-ready starting point.

The goal isn’t to replace developers, but to eliminate the repetitive setup work so builders can spend more time refining their ideas, polishing their projects, and creating a great demo.

Added proper guest mode to StartupOS! Users can now complete the founder interview, generate a blueprint, and explore the workspace without creating an account. Also fixed a bunch of authentication edge cases, improved session handling, and ran a full Playwright audit to make the app much more stable.

Thanks for the reviews, those helped really well. For some reasons my app broke during the QA Testing and I would apologize for that. (I have fixed it now hope so!). Also I have reworked the CSS(to make it less AI slop). Also this is 100% AI free. I hope you like it

StartupOS — The AI Operating System for FoundersInspirationStarting a startup is hard. Founders often spend weeks validating ideas, researching competitors, defining their target audience, creating business plans, and building an initial online presence.We wanted to build a platform that could compress this entire process into hours instead of weeks.That’s how StartupOS was born — an AI-powered operating system that helps founders go from a raw idea to a launch-ready startup.What it doesStartupOS acts as an AI co-founder that guides users through the earliest and most important stages of building a company.Users complete a guided founder interview, and StartupOS automatically generates:Comprehensive startup blueprintsBrand identitiesIdeal Customer Profiles (ICP)Revenue modelsProduct roadmapsCompetitor intelligenceDaily founder briefsAI startup roasts and feedbackLaunch-ready websitesEverything is managed through a unified founder workspace.How we built itStartupOS is built as a modern full-stack monorepo.FrontendNext.jsReactTypeScriptTailwind CSSshadcn/uiBackendFastifyTypeScriptPrisma ORMPostgreSQLAI InfrastructureMulti-provider AI routingGoogle GeminiGroqOpenRouterNVIDIA NIMDeploymentVercelRailwayDockerThe monorepo architecture allows the frontend, backend, and shared packages to evolve together while maintaining a clean developer experience.Challenges we ran intoBuilding StartupOS involved solving numerous engineering challenges:Designing a reliable multi-provider AI failover systemHandling provider outages and rate limits gracefullyMigrating from a monolithic architecture to dedicated frontend and backend servicesDebugging complex deployment issues across Railway and VercelImplementing persistent AI-generated assets and state managementCreating robust error handling and production-grade fallback systemsMaintaining a smooth user experience despite long-running AI tasksAccomplishments that we’re proud ofSuccessfully shipped a production-ready full-stack AI SaaS platformBuilt a complete founder workflow from idea to launchImplemented resilient AI failover across multiple providersDeployed scalable frontend and backend infrastructureAchieved successful end-to-end testing across the entire platformCreated a responsive experience across desktop, tablet, and mobile devicesWhat we learnedThrough StartupOS, we learned a tremendous amount about:Full-stack architectureProduction deploymentsAI orchestrationAPI reliabilityError recovery strategiesMonorepo development workflowsBuilding products for real usersWhat’s next for StartupOSOur roadmap includes:Visual website editorOne-click domain setupTeam collaboration featuresStartup analytics and trackingPDF and pitch deck exportsAI growth recommendationsMarketplace integrationsAdditional deployment providersStartupOS is just getting started. Our vision is to build the ultimate operating system for founders — an AI platform that helps anyone transform ideas into successful startups.

🚀 StartupOS V2 — From Idea to Launch, Powered by AIStartupOS V2 is the next generation of StartupOS, an AI-powered platform designed to help founders turn startup ideas into launch-ready businesses.This repository originally started as StartupOS V1, which contained both the frontend and backend in a single codebase. As the project grew, the backend was extracted into a dedicated repository to improve scalability, deployment, and maintainability, allowing the frontend to evolve independently as StartupOS V2.✨ FeaturesAI-powered founder interview flowAutomatic startup blueprint generationCompetitor intelligence dashboardDaily founder briefs and insightsWebsite generation powered by multiple AI providersInteractive founder workspaceBrand, ICP, revenue, roadmap, and startup roast tabsAuthentication with persistent sessionsFully responsive UI across desktop, tablet, and mobile🛠 Tech StackNext.jsTypeScriptTailwind CSSReactshadcn/uiVercelFramer Motion📈 EvolutionStartupOS V1Monolithic architectureFrontend + Backend in one repositoryStartupOS V2Dedicated frontend repositorySeparate production backend serviceImproved scalability and deployment workflowsBetter developer experienceProduction-ready architecture🎯 GoalStartupOS aims to dramatically reduce the time required to validate and launch startups by giving founders an AI copilot that guides them from raw idea to execution.Built with countless iterations, debugging sessions, deployments, and QA cycles — and now officially beta-ready 🚀#StartupOS #HackClub #Stardance #NextJS #TypeScript #TailwindCSS #AI #WebDevelopment #Startup

🚀 Shipped a major milestone for StartupOS Backend today!StartupOS Backend powers the entire StartupOS platform — an AI startup copilot that helps founders go from idea → blueprint → launch.What I worked on:✅ Fixed production deployment issues on Railway✅ Added robust AI provider failover across multiple providers✅ Implemented persistent website generation and retrieval APIs✅ Added competitor intelligence and daily brief endpoints✅ Improved error handling and graceful fallbacks across services✅ Fixed authentication/session issues and added token refresh support✅ Added production logging and route debugging utilities✅ Removed Redis dependency from critical flows to improve reliability✅ Performed full end-to-end QA on productionTech Stack:FastifyTypeScriptPrisma ORMPostgreSQLRailwayDockerMulti-provider AI Router (Gemini, Groq, OpenRouter, NIM)Current Status:🎉 Backend is now production-ready for beta users.All core flows are working:AuthenticationStartup creationBlueprint generationWebsite generationCompetitor intelligenceDaily briefsDashboard APIsReady for real user testing 🚀#Stardance #HackClub #TypeScript #Fastify #Prisma #PostgreSQL #AI #WebDev #StartupOS

🚀 Big StartupOS milestone today!After rebuilding the backend and fixing countless issues, StartupOS can now:✅ Generate startup blueprints✅ Generate actual multi-page websites with AI✅ Validate the generated HTML✅ Deploy websites to VercelThe journey now looks like:Idea → Blueprint → Website → Deploy → Public URLFeels amazing seeing StartupOS go from generating plans to generating real websites. 🎉Hours today: 5

Fixing website generations!
AI providors keeps failing

Worked on StartupOS today 🚀

✅ Improved AI website generation flow
✅ Fixed UI bugs and polished the dashboard
✅ Refined prompt engineering for better startup and business plan generation
✅ Tested deployment and overall user experience

Today’s focus was making StartupOS more reliable and easier to use. Still lots to build, but the product is starting to come together.

GitHub progress, testing, debugging, and feature development completed.