From Flying Taxis to Smart Steel: How Indian Startups Are Embedding AI into Core Industry Infrastructure
For the past two years, the conversation around artificial intelligence in India has been dominated by pilots: proof-of-concept projects, experimental deployments, and cautious initial rollouts. Enterprises dipped their toes into AI, but few had fully committed to making it core to their operations.
That era is ending.
Across aviation, enterprise software, and industrial automation, Indian startups and tech firms are rapidly moving beyond experimentation to large-scale AI deployment. Instead of remaining in the realm of isolated pilots, AI is now becoming core business infrastructure—embedded into the design of aircraft, the architecture of enterprise platforms, and the control systems of steel plants.
This shift is not incremental; it is foundational. And it signals that India’s AI ecosystem is maturing from a collection of point solutions into a deep-tech powerhouse reshaping entire industries .
✈️ Aviation: Digital Twins and Flying Taxis Take Off
The aviation sector, traditionally conservative and regulation-heavy, is witnessing some of the most ambitious AI deployments. At the forefront is The ePlane Company, a Chennai-based startup incubated at IIT Madras, which is developing India’s first electric air taxi, the e200x .
How AI Is Powering the e200x:
The startup has partnered with NVIDIA to create a high-fidelity digital twin of the aircraft using NVIDIA Omniverse libraries . This virtual replica allows engineers to:
- Simulate complex flight physics and aerodynamic interactions
- Test sensor performance and autonomy algorithms
- Run emergency scenarios and extreme weather conditions
- Log millions of virtual flight kilometres before physical tests begin
Bakthakolahalan Shyamsundar, Principal Engineer for Avionics Systems and Autonomy at ePlane, explained the significance: “The virtual testing environment allows the aircraft to undergo extensive learning before physical deployment. This approach enables engineers to explore rare scenarios, refine algorithms and validate decisions in a digital setting so that safety remains uncompromised in actual flight operations” .
The e200x will also integrate NVIDIA IGX as its onboard computing platform, processing continuous streams of data from cameras and radars to enable real-time decision-making . Crucially, the digital twin will serve as a predictive maintenance engine, mirroring aircraft components to detect potential issues early and reduce operational risks .
Satya Chakravarthy, Founder and CTO of ePlane, emphasized the broader vision: “We are not just building an aircraft; we are building an ecosystem. Collaborating with NVIDIA allows us to blur the line between the digital and the physical. By validating our flight operations suite in NVIDIA Omniverse, we are effectively pushing the limits of the aircraft thousands of times in simulation so that we never have to in reality” .
The first prototype has already been built, with ground testing scheduled to begin shortly. The company plans to build two more prototypes and pursue certification from the Directorate General of Civil Aviation (DGCA) before commencing operations in metropolitan cities such as Bengaluru, Mumbai, and Chennai .
This project marks India’s first use of NVIDIA Omniverse for multi-physics digital twin modelling by an electric aviation original equipment manufacturer, positioning the country among a select group of nations developing next-generation vertical flight systems at an industrial scale .
🏢 Enterprise Platforms: From Code Generation to Engineering Discipline
In the enterprise software space, the shift from experimental AI to core infrastructure is equally pronounced. Two recent launches exemplify this trend.
Kaara.Code: AI-Native Builder for Enterprise Delivery
On April 17, 2026, Kaara, an AI-first engineering and enterprise transformation company, announced the launch of Kaara.Code, an AI-native builder platform designed to address persistent challenges in enterprise technology delivery .
Over the last 13 years, Kaara has delivered more than 100 projects across industries including BFSI, healthcare, manufacturing, retail, and supply chain. Through these engagements, the company identified recurring delivery issues that continue to slow enterprise teams down—even as AI tools make code generation faster . These include repeated rebuilding of business context across initiatives, downstream bottlenecks in review and integration, and the growing sprawl of disconnected AI initiatives.
Kaara.Code is structured around three core layers :
| Layer | Function |
|---|---|
| Blueprint Layer | Captures enterprise context before development begins—business workflows, architecture, compliance needs, integration patterns |
| Mastery Layer | Brings engineering judgment and delivery practices developed over years of enterprise work |
| Memory Layer | Retains learning across the software development lifecycle, reducing repeated rediscovery |
Ashwini Suman, CEO of Kaara, articulated the philosophy: “There is a lot of conversation today around the role of software developer becoming obsolete because AI can generate code. We do not see it that way. AI is making some parts of development faster, but enterprise delivery still depends on engineering judgment, system understanding, architecture, integration, governance, and coordination. The developer’s role is not disappearing. It is evolving, and in many ways becoming even more important” .
Deloitte’s GenW.AI: Fully India-Developed Enterprise AI Platform
On February 10, 2026, Deloitte announced GenW.AI, its next-generation open-source, Java-based, low-code enterprise platform—the first of its kind developed entirely in India .
The platform helps organisations rapidly build applications, dashboards, and AI agents, and converge AI, Gen AI, and agentic AI under a single unified framework. It is designed to integrate with a wide range of large language models, ensuring flexibility as sovereign and enterprise-grade AI ecosystems evolve .
Key features include:
- 50% cheaper than competing market offerings
- Available both as on-premise and on-cloud, ensuring complete control over IP and data
- Integration of data scattered across functions and workflows
- AI explainability as a core design principle
Notably, of Deloitte India’s 45,000 employees, some 30,000—including 450 practice leaders—are already trained for the platform’s rollout. The platform has been tested internally across functions in India over the preceding six months .
Jagadish Bhandarkar, Partner and Chief Disruption Officer at Deloitte India, noted: “Enterprises today don’t just need tools. They need frameworks that let them move fast without increasing operational risk. The challenge isn’t about whether to adopt low-code or AI, but how to do so with guardrails, scale, and speed” .
⚙️ Industrial Automation: Steel, AI, and the 400 Million Tonne Target
Perhaps the most dramatic demonstration of AI moving from pilot to production is in India’s steel sector. At the India AI Impact Summit 2026 in New Delhi, the Ministry of Steel unveiled a comprehensive roadmap for digital transformation across the steel value chain .
The Scale of Ambition:
| Target | Timeline |
|---|---|
| Current crude steel capacity | ~200 million tonnes |
| Target capacity by 2030-31 | 300 million tonnes |
| Target capacity by 2035-36 | 400 million tonnes |
Steel Secretary Sandeep Poundrik highlighted that India’s steel consumption has nearly doubled from 77 million tonnes in 2014-15 to 152 million tonnes in 2024-25, reflecting rapid infrastructure expansion, urbanization, and manufacturing growth .
The AI in Steel Pavilion: From Pilots to Mission Mode
At the core of the initiative is the AI in Steel Pavilion, a first-of-its-kind collaborative platform that showcases real-time industry problem statements and invites AI solution providers, startups, and research institutions to co-create practical and scalable solutions .
Unlike conventional exhibitions, the Pavilion is designed as a problem-to-solution marketplace. It presents specific operational, logistical, safety, quality control, sustainability, and marketing challenges faced by steel producers and mining companies. The initiative signals a clear transition from incremental digitisation efforts to mission-mode AI integration across mining, logistics, production, quality assurance, marketing, and corporate governance .
High-Impact Use Cases Identified:
| Use Case | Description |
|---|---|
| Predictive maintenance | Reducing unplanned downtime in heavy machinery |
| Computer vision systems | Quality inspection and safety monitoring |
| Supply chain optimisation | Raw material blending and logistics efficiency |
| Intelligent decision support | Demand forecasting and yield improvement |
| Emission reduction | Lowering carbon footprint in steel production |
The Ministry has invited AI developers, deep-tech startups, academic institutions, and research laboratories to leverage the Steel Research and Technology Mission of India as a live sandbox environment .
The Secretary, Ministry of Steel, emphasized: “As capacity expands, productivity, quality, safety, and sustainability must improve proportionately. Intelligent automation, digital twins, advanced analytics, and AI-driven process control systems will be critical to ensuring that India’s steel growth remains globally competitive and environmentally responsible” .
🏗️ The Enabling Infrastructure: Reliance’s ₹10 Lakh Crore Bet
These deployments are being enabled by massive investments in AI infrastructure. At the same India AI Impact Summit 2026, Mukesh Ambani, Chairman and Managing Director of Reliance Industries, announced that Reliance and Jio will invest ₹10 lakh crore over the next seven years to build sovereign AI infrastructure in India .
Key Announcements:
| Initiative | Details |
|---|---|
| Jio AI Bharat | Multilingual AI capability across all Indian languages |
| Gigawatt-scale data centres | Construction started at Jamnagar; 120+ MW online by H2 2026 |
| Green energy advantage | Up to 10 GW of ready green power surplus |
| Nationwide edge compute layer | Deeply integrated with Jio’s network for low-latency intelligence |
Ambani stated: “Jio connected India to the internet era. Jio will now connect India to the intelligence era. We will deliver intelligence to every citizen, every sector of the economy, every facet of social development, and every service of government” .
He also articulated five non-negotiable principles guiding Jio Intelligence, including that “India cannot afford to rent intelligence. Therefore, we will reduce the cost of intelligence as dramatically as we did the cost of data” .
The government is complementing these private investments with policy support. Budget 2026-27 proposed a tax holiday till 2047 for eligible foreign cloud service providers operating through India-based data centres, providing long-term policy certainty for capital-intensive infrastructure projects .
Why This Shift Matters
The transition from experimental pilots to core infrastructure deployment carries profound implications for India’s startup ecosystem:
1. AI Is Becoming a Revenue Driver, Not a Cost Centre
When AI is embedded into core operations—whether in aircraft design, enterprise platforms, or steel production—it directly impacts the bottom line. This changes the ROI calculation for enterprises and makes AI investment a strategic imperative rather than an experimental luxury.
2. Scale Requires Integrated Solutions
As the Ministry of Steel’s roadmap demonstrates, moving from pilot to production requires solving integration challenges, scalability issues, and workforce adoption barriers. This creates opportunities for startups that can deliver end-to-end solutions, not just point products.
3. Industry-Academia Collaboration Is Accelerating
The ePlane Company’s IIT Madras incubation, the Ministry of Steel’s sandbox environment, and Deloitte’s internal training programmes all point to a deepening collaboration between research institutions, enterprises, and startups.
4. India Is Building Sovereign AI Capabilities
From Jio’s ₹10 lakh crore investment to Deloitte’s India-developed GenW.AI platform, there is a clear push toward building AI infrastructure and applications that are designed, developed, and deployed in India—not just imported and adapted.
5. The Role of the Developer Is Evolving, Not Ending
As Kaara’s CEO noted, AI is making some parts of development faster, but enterprise delivery still depends on engineering judgment, system understanding, architecture, integration, governance, and coordination. The developer’s role is becoming more important, not less .
The Road Ahead
The evidence is clear: Indian startups and tech firms are moving beyond AI experimentation to large-scale deployment. The ePlane Company is building India’s first electric air taxi using digital twin technology. Kaara and Deloitte are launching enterprise platforms designed for production-scale AI integration. The Ministry of Steel is driving AI-led transformation across the steel value chain. And Reliance Jio is investing ₹10 lakh crore in the infrastructure that will power it all.
This is not a future trend. It is happening now. And it signals that India’s AI ecosystem is entering a new phase—one where AI is not a separate “thing” to be adopted, but the core infrastructure upon which industries are built.
As the Secretary, Ministry of Steel, noted: “The next era of Indian steel will be shaped as much by data as by metallurgy” . The same can be said for aviation, enterprise software, and every other industry where Indian startups are embedding AI into the core fabric of operations.
The pilots are over. Production has begun.
