Projects

Financial planning in India is fragmented across banks, insurance, and tax documents. Most tools require sharing sensitive data with third parties.

Indian property buyers lose lakhs to fraudulent listings on Magicbricks, 99acres, Housing.com, and NoBroker. Fake RERA numbers, recycled stock photos, and inflated pricing slip past buyers because no neutral tool exists to verify a listing in seconds.

Business users need to query databases without knowing SQL. Existing tools lack context-aware query generation and safety guarantees.

India has 12 lakh+ small clinics running on phone calls and paper diaries. Patients call multiple times to confirm, double bookings happen daily, and revenue leaks through manual invoicing.

Travel apps optimize for proximity and ratings. They don't consider human energy levels, group dynamics, or intelligent budget allocation.

Devotional users in India juggle 5+ separate apps for Chalisa, Gita, Aarti, Ramayan, and Mahabharat. Most are ad-heavy, poorly designed, and lack multi-language support.

Private deal networks rely on manual introductions and spreadsheets. Matching the right profile to the right deal is slow, subjective, and misses opportunities.

Indian middle-class families track expenses across UPI apps, bank statements, and paper notebooks. No single tool auto-imports bank SMS, categorizes spending, and provides actionable savings advice — without cloud dependency.

Indian kirana stores and small retailers use paper registers or basic billing software with no inventory tracking, no GST compliance, and no offline support. Enterprise POS systems are too expensive and complex.

Group trip planning is fragmented across WhatsApp, Google Docs, Maps, Splitwise, and email. Nothing talks to each other, and every app breaks the moment you lose WiFi — exactly when you need it most.

Indians lose thousands of crores annually to digital scams via WhatsApp, SMS, and social media. Existing solutions require internet, collect user data, or lack Indian language support.

Middle-class families worldwide track expenses inconsistently — UPI apps show transactions but don't enforce budgets. No tool combines expense tracking, grocery budgeting, bill calendars, debt management, and family splitting in one offline app.

Engineering work is often invisible. Portfolios show polished results but not the thinking behind them.

Most AI agent failures are silent — the run completes, the status code is green, and the result is still wrong. Observability platforms show you the trace, but never the cause of death.

Training your own language model sounds like a PhD job that needs a GPU cluster, so most developers never try. The tooling looks intimidating from the outside.

A photo shows you one side of an object; the other five sides are a guess. Turning a single 2D image into a usable 3D model normally needs a photogrammetry rig or hours of manual modeling.

An agent setup grows one plugin at a time until it has hundreds of skills, and nobody can see what it actually contains. There is no view of which skills overlap, which clusters are overweight, or which ones were hand-written versus pulled in by a marketplace plugin.

Agentic coding tools stop at a diff. Nothing forces a spec to exist before implementation, nothing blocks a run that never wrote a test, and nothing catches a stage that quietly produced no artifact — so failures surface as a plausible-looking PR nobody can trust.

Farmer-facing AI tools die at the API-key step, answer in English, and hand back generic advice with no live numbers behind it. A smallholder asking when to sell or how to treat yellowing wheat leaves gets a paragraph, not a decision.

Restaurants run on 3-5% net margins while food costs are up roughly 35% since 2019, and most owners never see per-dish economics. The leaks — a bestseller that bleeds, a price that never moved when its ingredients did — hide inside a POS export nobody reads.

Most AI research tools are one model in a loop. It cannot hold a multi-hop question together, it has no second opinion, and vanilla vector RAG fumbles exactly the questions that need relationships rather than similar-looking paragraphs.

Regex is not hard because the syntax is exotic. It is hard because you can never be sure the pattern does what you think — research on regex use names trust, test coverage, and cross-language portability as the real pains. Most AI regex generators print a pattern and wish you luck.

Agent skills are the fastest way to teach a coding agent a new tool, but writing a good SKILL.md by hand means reading the documentation yourself and distilling it into trigger phrases, usage, recipes, and gotchas. Most people never get past the reading.

Restaurants throw away 4-10% of the food they buy before it reaches a plate, and most owners find out at month end from a food-cost percentage that moved the wrong way, with no idea which ingredient did it. The tools that solve this sit behind a per-location subscription, a sales call, and your data on someone else's servers.

Home design tools want an install, a login, or a CAD background. Someone who just wants to see whether a sofa fits has no fast way to lay out a room and look at it.

Voxel sandboxes are the demo everyone builds with a coding agent right now, and almost all of them are a throwaway single HTML file that hitches on chunk generation, ships a texture pack, and cannot be read or extended.

Generative interfaces usually hand a model the wheel and hope. Let an LLM write shader code or scene state directly and one hallucination breaks the render — which is why most text-to-3D demos snap between presets instead of responding continuously.

Hexapod gait and leg IK are usually explained with equations and a video, or hidden inside a robotics library. Neither lets you feel what changing duty factor or step height actually does to a walking robot.

Face and hand tracking demos come wrapped in a build step, a server, and usually an API key — which puts a wall in front of anyone who just wants to see what the models actually output before building on them.