AI Development Projects in KarnatakaAI Development Projects in Karnataka
Cover image for Take back the control of your digital life. Built with @Anyt...
Take back the control of your digital life. Built with @Anything Ai
5
14
1.1K
Cover image for WrkOS One Workspace. Multiple Products.
A
WrkOS One Workspace. Multiple Products. A scalable SaaS platform combining documentation, CMS, collaboration, publishing, and organization management in one workspace.
0
317
Cover image for Something I’ve learned working with clients: They rarely nee...
Something I’ve learned working with clients: They rarely need “fancy.” They need results. Two weeks ago, a client came to us overwhelmed by their data. They had sales data scattered across spreadsheets, apps, screenshots, and random exports. They were making decisions based on guesses, not facts. We built them an e-commerce sales dashboard that cuts straight to what matters — revenue trends, top products, customer behavior, and real conversions. And here’s the lesson: Most businesses aren’t struggling because they lack data. They’re struggling because they can’t understand it. Once you turn information into clarity, everything else becomes easier. And we’re not done yet. AI-driven sales insights are coming next. Stay tuned.
0
246
Cover image for An Interactive 3D Isometric Archive
An Interactive 3D Isometric Archive (2018–2026) 💿✨ Live App: https://chronos-motion-scroll.base44.app (https://chronos-motion-scroll.base44.app)The Vision: Making Digital Memory Tactile Most portfolio timelines and personal archives on the web feel lifeless, flat 2D sliders, endless vertical scrolls, and static grids. For the #Base44Challenge, I wanted to turn a personal timeline spanning 2018 to 2026 into a tactile, physical experience: like rifling through a crate of vintage CD jewel cases or vinyl records on a studio desk.
1
2
236
Cover image for CrawlX
CrawlX crawls a website, surfaces
CrawlX CrawlX crawls a website, surfaces its technical SEO problems, and explains each one in plain language with a suggested fix. Cloud-native, fast, and built for teams. I built it from a blank page: product, brand, and engineering. The gap The technical SEO crawler market is led by Screaming Frog, a desktop tool with a dated interface, no AI, and a one-person, one-machine model. No sharing, no assigning issues, no client portals. At the other end, enterprise crawlers like Lumar are fast but cost $15k to $50k a year. That left a clear gap: enterprise speed, a modern interface, AI built in, and pricing a small team can actually afford. Positioning Linear meets Datadog for SEO. A crawler that does not just hand you a list of errors. It tells you what each one means and how to fix it. What I built 1. Product, engineering, and brand, end to end. 2. 400+ SEO checks across 12 analysis modules 3. An AI layer on the Claude API: plain-English explanations, auto-generated fix code, and a conversational assistant you can ask about your own crawl 4. Six core screens, including a live crawl view, a Kanban issue explorer, a site-architecture visualizer, and a white-label report builder 5. An AI search-readiness module that checks whether pages are built for AI answer engines, a gap no competitor covers 6. Rust crawl workers hitting 350 to 450 URLs per second, matching enterprise tools, with ClickHouse handling billion-row crawl data What makes it different 1. AI intelligence no legacy crawler offers: explanations, fix code, and a chat assistant, not static hints 2. Enterprise-class crawl speed at small-team pricing 3. Team-first: assign issues to developers, share client portals, comment in place, instead of one license per machine 4. Built into the pipeline: CI/CD on GitHub Actions and Vercel catches SEO regressions before they ship
0
176
Cover image for Teams running LLM features in
Teams running LLM features in production watch the bill climb and can't say why. Which feature is driving it? Which model? How much goes to retries? Most optimisation advice is a percentage nobody actually measured. llm-audit is a command-line toolkit I built to answer those questions with evidence. HOW IT WORKS → Ingest: OpenTelemetry GenAI traces, Langfuse data, or usage exports from OpenAI and Anthropic, loaded into one DuckDB table. → Analyze: 16 deterministic checks, 8 for cost (unused prompt cache, model overkill, duplicate calls, context bloat, retry waste, off-peak eligible work and more) and 8 for reliability (error rates, latency, rate limiting, single-provider risk and more). → Backtest: replays the workload with each fix switched on alone, then compares predicted and measured savings. → Report: one HTML file with ranked findings. Each finding has its evidence query, a fix and a confidence level. WHAT MAKES IT DIFFERENT • Content-free. It works from prompt hashes and token counts only, so prompt text never has to leave the client. • No unmeasured claims. A saving is either measured by the backtest or labelled as an estimate. A fix that lowers quality is marked rejected, even when it saves money. • Honest gaps. Checks that can't run on the data are listed with the reason, not quietly dropped. THE PART I'M PROUDEST OF My first paid backtest failed its own checks: 75% recall and 103% savings error. Instead of tuning the numbers, I debugged the harness. The test runs were sharing the provider's prompt cache, so they weren't independent. A fresh marker per run fixed it, and the re-run confirmed the isolation (0.0 cache hits where there had been up to 1.0). FIELD-TESTED ON MY OWN APP I audited AI Lab, my public Q&A app over the EU AI Act and GDPR. Finding zero: the app kept no per-call record at all. My merged PR added a content-free, per-attempt trace. The audit then showed that the evaluation judges took about two thirds of the spend, so scoring answers cost twice as much as producing them. BUILT LIKE THE AUDIT IT SELLS 915 tests passing · dependencies pinned with hashes · $0.97 total LLM spend for the whole build, under a $15 cap enforced in code. The demo workload is synthetic and the field run covers one app, so no percentage here applies to your bill. An audit of your own data is the only way to find out. A good starting point is a Bill Snapshot: send your usage exports and get ranked findings back in 48 hours. #LLMOps #FinOps #AIEngineering #GenAI #LLM #AICostOptimization #Observability #OpenTelemetry #Python #DuckDB #AIAudit #PrivacyByDesign
0
133
Cover image for For this cinematic sci-fi book
For this cinematic sci-fi book trailer, I transformed the story into a visual narrative by first breaking the novel into key emotional moments and writing a trailer script. I then created consistent character designs and generated AI-powered imagery and video sequences to capture the loneliness of deep space and the emotional weight of a mother's journey home. Finally, I edited the scenes together with cinematic music, sound design, voiceover, and pacing techniques to create a compelling trailer that highlights the book's themes of love, sacrifice, and time.
2
2
308
Cover image for Hair AI - Privacy-First B2B
Hair AI - Privacy-First B2B Beauty Platform with Zero API Costs
0
91