The InBharat.ai ecosystem is a portfolio of twelve interconnected AI products—agentic search, interactive storybooks, a company OS, a personal AI OS, a field health app, education tools, a WhatsApp bot, decision intelligence, agent observability, social automation, and a developer installer—built for India because Indian problems are interlocked: a farmer needs credit, market info, and learning together. One product can’t serve those needs; twelve that share an identity and data layer can.
A practical, jargon-free guide for Indian engineering teams and founders — part of the Learn AI with Reeturaj series on InBharat AI.
I get asked a lot: what's the strategy? Why twelve different products? Why not focus on one thing?
The answer is India. India's problems aren't one thing. They're interconnected. They're systemic. You can't solve one without touching the others. So we've built not a product company. We've built an ecosystem.
Let me walk through what we're building and why the portfolio makes sense.
There's a way that American tech companies think about products. Find a problem. Solve that problem really well. Get product-market fit. Scale. Go public. Become a trillion-dollar company.
That works when you're solving a problem for wealthy people who have money to spend, stable internet, formal education, existing financial infrastructure.
It doesn't work the same way in India.
In India, you need to think about problems ecologically. The problems are interlocked. The solutions are interlocked. The users are the same people. You can't serve a farmer with just one product because a farmer's needs span multiple domains.
I'll walk through what we're building, but I want you to understand the connective tissue. How they work together.
Agentic search and voice: InBharat AI is the flagship—a multi-agent agentic search and chat platform, voice-first and multilingual, with specialist agents for coding, education, research, shopping, and enterprise. It's the front door. Sahaayak AI extends that idea into a multilingual personal AI OS for 1.4B Indians—chat, voice, OCR, translation, smart notes, email intelligence, live news. UniBot meets people where they already are: a conversational AI support bot on WhatsApp, multilingual, for helpdesk and guidance.
Education: KathaKitaab turns Indian epics into living, interactive storybooks with verb-aware AI animation and Indian-language narration. UniAssist.ai guides students through international admissions—university matching, PR prediction, scholarship discovery. TestsPrep.in does adaptive exam preparation—practice tests, performance analytics, focused pathways. Three products, one learner, from childhood stories to competitive exams.
Public services and health: SahaayakSeva is a field AI app for India's Anganwadi workers—WHO-standard child growth tracking, GPT-4o Vision OCR, maternal risk scoring, and government scheme lookup. It's where AI meets the last mile of public health.
Enterprise and agents: JAK Swarm is an open-source, self-hostable closed-loop company OS—evidence graph, drift detection, executable specs, and risk-gated agent approvals with audit trails. Agent Arcade is the observability layer—a live dashboard, session replay, and cost analytics across 29 models. Together they let a company run AI agents with proof.
Decision intelligence: Phoring turns documents into a knowledge graph, runs multi-agent simulations, and returns source-cited forecasts. It's the thinking tool for people who need to reason about outcomes, not just retrieve them.
Builder tools: SocialFlow is open-source AI social-media automation—generate and publish to 12 platforms, with AES-256 local credentials so your data never leaves your machine. OpenClawFix is a one-click installer for the OpenClaw IDE via Docker, auto-configuring gateway tokens and 16 AI providers. These are the tools that make the rest faster to build and ship.
Twelve products. One identity layer. One data layer. One mission.
A farmer doesn't just need agricultural AI. They need to understand financial products to get credit for seeds. They need to access market information. They need to learn new techniques. They need to transact money. A single app can't carry all of that—but a connected ecosystem can, and the same user moves across it.
A student doesn't just need test prep. They need stories that build language early. They need guidance on where to study. They need a bot that answers questions on WhatsApp at midnight. KathaKitaab, UniAssist.ai, TestsPrep.in, and UniBot serve one learner across a decade.
A frontline health worker needs diagnostic support, record sharing across facilities, and education to stay current. SahaayakSeva and the education products meet them together.
One product doesn't serve these needs. But twelve products that work together do.
This only works if the products actually connect. If they share data intelligently. If they understand each other.
We're building a shared identity layer on top of Aadhaar. You're one user across all twelve products. Your learning data connects to your search history. Your health records connect to your education. Your commerce activity informs what the agents recommend.
This is privacy-preserving. This is consent-based. But it's the connective tissue that makes the ecosystem work.
We're building a shared data layer. Language models trained across the full scope of Indian life, not narrow domains. So the search agent understands the education context. The health app understands the local-language context. The models learn from the whole, not the slice.
We're building shared infrastructure. Local-first infrastructure. Data stays in India. Processing happens in India. The twelve products run on the same core, which means we can operate at a scale that one product never could.
I think about this ecosystem approach as the only way to compete globally while serving India.
If you build one product and try to go global, you get absorbed into the American tech industry. You compete with American companies at their game. You lose.
But if you build an ecosystem of products specifically for India, serving India deeply, you create something that's defensible. That's valuable. That's hard to compete with because it's built for problems that American companies don't understand.
Alibaba didn't win by building a shopping app. It won by building an ecosystem. Finance, cloud, logistics, education. All interconnected. All serving China. That's the model. That's what we're building for India.
We're past vision and into shipping. InBharat AI is live—agentic search, voice, and the specialist agents are working. KathaKitaab is live with interactive storybooks in Indian languages. JAK Swarm is open-source and self-hostable today, with the evidence graph and JAK Shield risk gating. SahaayakSeva is in the field with Anganwadi workers. UniAssist.ai, TestsPrep.in, and UniBot are serving students. Phoring, Agent Arcade, SocialFlow, and OpenClawFix are the builder layer, all usable now.
This isn't a five-year vision. This is being built right now.
If you're thinking about joining us, understand what you're joining. You're not joining a company with one product and one customer problem. You're joining an effort to build the foundational AI infrastructure for India.
You'll work on problems that American tech companies have never solved. You'll work at scale. You'll work on infrastructure that matters. That impacts hundreds of millions of people.
And you'll own something. You'll build something that's India's. That works for India. That's made in India.
I believe India will be built on Indian infrastructure. Indian companies. Indian talent. Not because of nationalism. Because it's the only way to solve Indian problems well. Because it's the only way to reach India's 500 million unconnected people. Because it's the only way to build something that's defensible.
Twelve products. One mission. Build AI for Bharat.
Why build twelve products instead of focusing on one? India's problems are interconnected—a farmer needs credit, market access, and learning together; a student needs stories, guidance, test prep, and a bot that answers at midnight. Serving one need in isolation doesn't work in India. An ecosystem of connected products serves the same user across a decade of needs.
What are the twelve products in the InBharat ecosystem? InBharat AI (agentic search), KathaKitaab (interactive storybooks), JAK Swarm (company OS), Sahaayak AI (personal AI OS), SahaayakSeva (field health app), UniAssist.ai (admissions guidance), TestsPrep.in (exam prep), UniBot (WhatsApp bot), Phoring (decision intelligence), Agent Arcade (agent observability), SocialFlow (social automation), and OpenClawFix (developer installer)—all built for Indian context and constraints.
How do the products connect? A shared, consent-based identity layer (one user across all products), a shared data layer where models learn across the full scope of Indian life, and shared local-first infrastructure—data and processing stay in India—so the search agent understands the education context, the health app understands the local language, and so on.
Isn't a multi-product approach unfocused? It's the defensible model. A single product going global gets absorbed competing with American companies on their terms. An ecosystem built deeply for India—like Alibaba for China—is hard to compete with because it's built for problems American companies don't understand.
What's the current state of the ecosystem? Shipping, not vision—InBharat AI, KathaKitaab, JAK Swarm, SahaayakSeva, UniAssist.ai, TestsPrep.in, UniBot, Phoring, Agent Arcade, SocialFlow, and OpenClawFix are all live and usable now. It's being built right now, not on a five-year roadmap.
Reeturaj Goswami is the founder of InBharat.ai, building an ecosystem of AI products specifically for India's unique challenges and opportunities.
#InBharat #DeshKaAI #AIForBharat #IndianStartup #AIEcosystem #BuildingBharat