top of page

The Start-up Building a Star in Bengaluru

Apr 26
11 min read

By Shivansh Rana


 Aerospace Engineering, Amity University


 Published: April 2026 | Category: Energy

1. The Energy Question:

By 2040, India's electricity demand will roughly double. In practical terms, the country must build a second national grid equal to what it took a century to create in under twenty years. The numbers are already moving. India adds 4 to 5 gigawatts of capacity each month. Total installed capacity has crossed 524 GW [1]. In 2024-25, the grid met a peak demand of nearly 250 GW a record unlikely to stand for long [2]. According to the International Energy and Climate Centre, no country will add more to global energy demand over the next two decades than India [3]. The energy transition is not a spectator sport here. It is being defined here. To its credit, India has moved fast. By January 2026, 52.3% of total power capacity came from non-fossil sources solar, wind, hydro, and nuclear [4]. This target was met five years ahead of the Paris Agreement schedule [5]. Solar alone has reached 140 GW, now the largest single contributor by capacity [4]. With ₹30 lakh crore in investment aimed at 500 GW of clean power by 2030, the direction seems clear [1]. But "capacity" is not the same as "electricity actually delivered." Despite clean sources accounting for over half of installed capacity, coal still supplies roughly 70% of the power that flows through the grid [6]. The reason is straightforward. Solar panels generate power less than a quarter of the time. Wind turbines fare only slightly better [6]. Coal and nuclear plants, by contrast, run reliably day and night. And peak demand arrives in the evening exactly when solar disappears [7]. The grid has no choice but to fall back on coal. This is not just an environmental problem. It is a strategic one. India imports nearly 90% of its crude oil, leaving the economy exposed to global price swings and geopolitical disruption [8]. Energy dependence is a vulnerability no amount of solar panels alone can fix. Batteries and storage will help. But storing enough power for 1.4 billion people through the night remains an enormous, unsolved challenge. This is why fusion matters. Unlike fission, which splits atoms and leaves behind long-lived waste, fusion joins them the same process that powers the Sun. It produces no carbon emissions. Its fuel comes from seawater. And unlike solar or wind, it runs around the clock. For decades, fusion belonged to massive government projects like ITER. That is changing. Private companies are now entering the race with smaller, faster approaches. And one of them is building its first machine in Bengaluru. If it works, if physics holds, fusion won't just be a scientific achievement. It will be the foundation of India's energy independence.

2. Who Has Tried? Who Is Trying?

For decades, fusion belonged to governments. The centerpiece is ITER in France, a $20 billion collaboration of 35 nations, including India, designed to achieve net energy gain for the first time [9]. National labs have pushed the boundaries: the UK's JET produced a record 59 megajoules of fusion energy [10]. South Korea's KSTAR sustained 100-million-degree plasma for 48 seconds [11]. China's EAST recently broke through a key plasma density limit [12]. These machines proved fusion works. They also proved it is slow and expensive. India has been present throughout. The Institute for Plasma Research (IPR) in Gandhinagar operates two indigenous tokamaks ADITYA-U and SST-1 and manages India's contributions to ITER, including the massive cryostat housing the reactor core [13]. The expertise is real. But it has always lived in the public sector. That is changing globally. Private startups are entering the race with faster, leaner approaches. Commonwealth Fusion Systems in the US has raised nearly $3 billion and is building SPARC, a compact tokamak using high-temperature superconducting magnets [14]. Helion has built seven prototypes, achieved private-sector deuterium-tritium fusion, and is constructing a plant for Microsoft [15]. Tokamak Energy in the UK and NovaFusionX in China are pursuing similar paths [16] [17]. Three things made this possible: High-Temperature Superconductors for stronger magnets in smaller machines, tungsten and advanced materials for surviving extreme heat, and AI simulation for testing designs virtually. Fusion has not become easy. It has become faster. For years, India watched this shift from the sidelines, its fusion expertise confined to government labs. That changed with the founding of Pranos Fusion in Bengaluru—India's first private venture to build a tokamak.




3. Enter Pranos Fusion 

In 2024, two engineers in Bengaluru concluded that India could no longer afford to stand apart from the private fusion race. Shaurya Kaushal, a physicist trained at the Jawaharlal Nehru Centre for Advanced Scientific Research, and Roshan George, previously chief engineer at REConnect Energy, established Pranos Fusion the country's first privately funded venture to attempt the construction of a tokamak [18]. What they were attempting was not incremental. It meant stepping into a difficult engineering problem and pursuing it without the scale or safety net of a government programme. The company benefits from institutional affiliations that grant it access to decades of accumulated expertise. Pranos is co-incubated at the Jawaharlal Nehru Centre for Advanced Scientific Research (JNCASR), operating under India's Department of Science and Technology, and at the Institute for Plasma Research (IPR) in Gandhinagar which is the nation's principal government laboratory for fusion research [18]. The enterprise also maintains formal ties to the ITER programme, placing it within the broader international fusion effort from its inception.

Pranos has not staked its future on a single technological breakthrough. Instead, the company is developing three interconnected systems in parallel [19]:

  • JENGA: A software platform for the design and control of tokamak plasmas. One may think of it as the operating system for a future fusion reactor.

  • PRAGYA: The physical apparatus a compact, low-aspect-ratio tokamak intended to study plasma behaviour and accumulate experimental data.

  • MAGGA: A dedicated programme to advance high-temperature superconducting (HTS) magnet technology, which permits stronger magnetic confinement within significantly smaller reactor volumes.

In March 2026, the company secured $6.8 million in seed funding. The round was co-led by pi Ventures and Ankur Capital [19]. The capital is directed toward a single, clearly defined objective: achieving first plasma in PRAGYA before the close of 2026. The term "first plasma" requires some clarification. It does not signify fusion. It does not indicate net energy production. It describes, quite precisely, the moment at which a tokamak succeeds in generating and magnetically confining a superheated, electrically charged gas plasma for the first time. The achievement is analogous to a newly constructed aircraft performing its initial taxi test: outwardly unremarkable, yet to the engineers who have brought the machine into being, indispensable. "Fusion has been proven scientifically," Kaushal has observed. "The real challenge now is making it commercially viable" [19]. The wager Pranos is making is that viability lies in smaller, more intelligent machines—compact tokamaks that use advanced magnets to deliver performance once thought to require vast installations. This brings us to PRAGYA. It is not a power station. It will never supply a single watt to the grid. Its value lies elsewhere. Before anyone can build a fusion power plant, they must first understand how plasma behaves inside a tokamak, how it moves, how it heats, how it escapes. Those answers cannot be found in theory alone. They must be measured. PRAGYA exists to take those measurements. If it functions as intended, it will produce something arguably more vital at this stage: reliable data, operational experience, and tangible proof that India can build tokamaks that compete. 

4. What exactly is PRAGYA?

At the centre of Pranos Fusion's effort lies PRAGYA a compact, low-aspect-ratio tokamak designed to confine and study superheated plasma. A tokamak, in essence, is a magnetic bottle. It uses powerful electromagnetic fields to hold a gas so hot millions of degrees that electrons are stripped from their atoms, forming an electrically charged plasma. This plasma must never touch the walls of the vessel; if it does, it cools instantly and the experiment fails. The magnetic field holds it suspended, like an invisible cradle.

Fig. 2 3D CAD model of the PRAGYA vacuum vessel (VV) assembly.

Fig. 2 Shows 3D CAD model of the PRAGYA vacuum vessel assembly. The vessel is divided into two electrically isolated halves (Torus 1 and Torus 2), separated by G10 insulation (visible as the thin layer between them). Multiple ports around the structure provide access for diagnostics, fuel injection, vacuum pumping, and viewing windows. The support legs and connecting trusses are also shown. 

The image above shows the vacuum vessel as a digital model, the blueprint from which the physical machine is now being built. The vessel is split into two halves, separated by a layer of electrical insulation. This gap called a toroidal electric break prevents unwanted currents from circulating through the metal walls, a common problem in conventional tokamaks that wastes energy and destabilizes the plasma. Numerous ports dot the surface viewing windows for cameras, rectangular openings for diagnostics and microwave injection, and smaller ports for fuel, vacuum pumping, and venting. Each must be sealed with precision to maintain the extreme vacuum inside. PRAGYA is small by the standards of government fusion programmes. Its plasma ring measures just 0.4 metres across, roughly the size of a large dinner plate. It is designed to carry 25,000 amperes of current and generate a magnetic field of 0.1 Tesla [18]. Its compact, spherical-like shape is a deliberate choice: such designs offer better stability than the larger and bulkier tokamaks of the past. Inside the vessel, conditions are extreme. The pressure must be reduced to one ten-billionth of normal atmospheric pressure, in case any leak would ruin the experiment. To guard against this, the joint where the two halves of the vessel meet is sealed with two O-rings, one behind the other. The space between them is pumped clear. If the outer ring fails, the inner ring holds [18]. The structural integrity of the vessel has been rigorously tested not on a physical prototype, but inside a computer. Using finite element analysis, the engineering team built a high-resolution digital replica of the vacuum vessel containing over one million mesh elements. They then subjected this virtual model to the loads it will experience in operation: the crushing pressure difference between the vacuum inside and the atmosphere outside, the weight of the vessel itself, and the thermal expansion that occurs when the entire structure is baked at 150 degrees Celsius for 48 hours. This baking process is essential; it drives out absorbed water and hydrogen from the steel walls that would otherwise contaminate the plasma during operation.

Fig. 3 Spatial distribution of temperature on the vacuum vessel due to vacuum, self-weight and baking at 150 ◦C.

Fig. 4 Spatial distribution of deformation (total) on the vacuum vessel due to vacuum, selfweight and baking at 150 ◦C. In (b), the deformation is scaled 100 times for the ease of visualization.

Fig. 3 shows the temperature distribution across the vacuum vessel during baking at 150°C. The interior surfaces are heated using infrared lamps, though some regions run slightly cooler due to heat loss and the vessel's geometry. Fig. 4 shows the resulting deformation under combined vacuum pressure, self-weight, and thermal expansion. The actual displacement is tiny, just 2.1 millimetres at the top of the vessel.

The story these images tell is straightforward. Heat is applied to the interior (Fig. 3), and the structure responds (Fig. 4). The temperature does not rise uniformly; some areas lag behind as heat conducts through the steel and escapes to the surrounding air. This uneven heating causes different parts of the vessel to expand at slightly different rates, producing the deformation pattern visible in Figure 4. In reality, the movement is barely noticeable. To make it visible, the engineers have exaggerated the deformation by a factor of 100 in the image. The support legs are designed with slightly oversized bolt holes, allowing the entire vessel to expand and contract gently as it heats and cools—without building up dangerous stress [18]. These simulations are not academic exercises. They are the evidence that convinced investors PRAGYA would survive. Before a single steel plate was cut, the design had passed every virtual test. The vessel will not buckle. It will not crack. It will not leak. PRAGYA is designed to answer a fundamental question: how does plasma behave inside a tokamak? Before any fusion power plant can be built, engineers must understand how plasma moves, heats, and escapes. Theory alone cannot provide those answers. They must be measured. PRAGYA exists to take those measurements, testing plasma configurations of varying shape, developing superconducting magnets and advanced control systems, investigating plasma stability, and trialling remote handling techniques essential for future reactors. If it functions as intended, PRAGYA will not generate electricity. But it will produce something equally vital: reliable data, operational experience, and the confidence to take the next step. Every large machine begins as a small one. PRAGYA is that first step for India's fusion journey, a step toward cleaner energy, greater self-reliance, and a place among the nations shaping the future of power.



5. The Road Ahead and Why It Matters

The first milestone is clear: by the close of 2026, PRAGYA must generate and magnetically confine a superheated plasma, proof that the design works beyond the simulation screen. Next comes PraniQ, a larger tokamak equipped with high-temperature superconducting magnets and designed to attempt net energy gain. Beyond PraniQ lies the commercial objective: a fusion plant generating 50 to 100 megawatts of clean, firm power for the grid. Shaurya Kaushal is candid about the timeline. Fusion power plants, he acknowledges, remain at least a decade away. In the near term, the company will generate revenue from its software platform, JENGA, and its superconducting magnet programme, MAGGA. Both technologies have immediate applications in medical imaging, defence, and transport, well before fusion electricity reaches the grid. Why does this matter for India? Electricity demand is projected to double by 2040. Renewables continue to expand, yet the requirement for firm, dispatchable power remains unmet. The SHANTI Act of 2025 has, for the first time, opened the nuclear sector to private participation, establishing a policy framework within which ventures such as Pranos may operate. Concurrently, India's contributions to ITER, the cryostat, cooling systems, and neutron shielding which has demonstrated that domestic industry already possesses the capacity to manufacture fusion-grade components [22]. India already possesses the necessary talent, an emerging supply chain, and an evolving policy framework. Pranos is not an isolated venture; it is the first visible expression of a broader national capability.  Zoom out further, and the Pranos story reflects a global shift. For decades, fusion belonged to governments—slow, expensive, and cautious. That is changing. Private companies across the United States, Europe, and China are now racing to shorten the timeline, betting that agile engineering and venture capital can succeed where large programmes have moved slowly. Through Pranos, India has entered that race. The outcome is not assured. Fusion remains among the hardest engineering challenges ever attempted. Yet the attempt itself—designing a tokamak, training engineers, developing magnets—builds capacity that serves the nation regardless of when fusion power reaches the grid.

Kaushal (Co-Founder, Pranos Fusion) puts it simply: "We stand on the shoulders of brilliant fusion physics. Now the world needs the commercial infrastructure to bring it to the grid. At Pranos, we are building exactly that, and we are beginning our contribution today, from India" [23].


JRC Takeaway:

  • Pranos Fusion is India's first privately funded tokamak venture, founded in Bengaluru in 2024.

  • PRAGYA is a compact test platform, not a power plant, designed to study plasma behaviour and validate engineering choices.

  • First plasma is targeted for late 2026. The design has passed rigorous virtual testing.

  • The roadmap extends to PraniQ (net energy attempt) and eventually a 50–100 MW commercial plant.

  • Fusion remains difficult. It is no longer slow. And India has just entered the race.





Refrences 

[1] CNBC TV18. "India adding 4–5 GW power capacity/month, to spend $3.03 trillion on smart meters for efficiency." January 14, 2026. https://www.cnbctv18.com/energy/india-adds-4-5-gw-power-capacity-monthly-plans-usd-3-03-trillion-push-on-smart-meters-ws-l-19821908.htm/amp

[2] Press Information Bureau, Government of India. "POWER SUPPLY, PEAK DEMAND AND AVAILABILITY OF COAL." March 30, 2026. https://www.pib.gov.in/PressReleaseIframePage.aspx?PRID=2246901

[3] EQ International. "India's energy consumption is expected to rise by 3-4 times over a decade: IECC." August 5, 2024. https://www.eqmagpro.com/indias-energy-consumption-is-expected-to-rise-by-3-4-times-over-a-decade-iecc-eq/

[4] Mercom India. "India's Non-Fossil Power Capacity Reaches 52.3% as of January 2026." March 18, 2026. https://www.mercomindia.com/indias-non-fossil-power-capacity-reaches-52-3-as-of-january-2026

[5] The Hindu. "How will India's new Nationally Determined Contribution accelerate climate action?" March 29, 2026. https://www.thehindu.com/sci-tech/energy-and-environment/how-will-indias-new-nationally-determined-contribution-accelerate-climate-action/article70797328.ece

[6] Ram Ratan, PhD. "India's Renewable Energy Boom: Challenges and Opportunities." LinkedIn. July 17, 2025. https://www.linkedin.com/posts/ramratan_renewableenergy-indiaenergy-energytransition-activity-7351550310269534211-TMra

[7] IAS Exam. "Share of Clean Energy in India's Electricity UPSC." July 18, 2025. https://www.iasexam.com/share-of-clean-energy-in-indias-electricity/

[8] 环球网 (Global Times). "提高本土产量，摆脱进口依赖，印度投资千亿美元扩产石油." January 2026. https://world.huanqiu.com/article/4QA1RcY40Nj

[9] 中国国际核聚变能源计划执行中心. "关于征集2026年ITER管理评估意向评估团队的通知." March 9, 2026. https://iterchina.cn/tzindex/info/2026/24220.html

[10] Van Eester, D. et al. "ICRH modelling of the Baseline D-T scenario in JET." EPJ Web of Conferences. January 7, 2026. https://epjwoc.epj.org/articles/epjconf/abs/2026/02/epjconf_rfppc2026_02027/epjconf_rfppc2026_02027.html

[12] Chinese Academy of Sciences. "EAST Tokamak Experiments Exceed Plasma Density Limit." February 10, 2026. http://english.cas.ac.cn/newsroom/research-news/202602/t20260210_1150309.shtml

[13] Facilitation Centre for Industrial Plasma Technologies. "Institute for Plasma Research." http://plasmaindia.com/

[14] Nuclear Engineering International. "AI digital twin for SPARC." January 8, 2026. https://www.neimagazine.com/news/ai-digital-twin-for-sparc/

[15] Helion Energy. "Helion Achieves New Industry-first Fusion Energy Milestones." February 13, 2026. https://www.helionenergy.com/articles/helion-achieves-new-fusion-energy-milestones/

[16] GOV.UK. "Magnets Partner Agreement - UK Fusion Energy Ltd." March 25, 2026. https://www.find-tender.service.gov.uk/Notice/027449-2026

[17] Shanghai Municipal Information Office. "NovaFusionX sets fundraising record for a private China fusion startup." April 12, 2026. https://touch.shio.gov.cn/jsp/jjxw_detail_en.jsp?id=20260412191160080

[18] Pranos Fusion et al. "Design and mechanical analysis of the PRAGYA tokamak vacuum vessel." arXiv:2603.11549. March 2026. https://browse-export.arxiv.org/abs/2603.11549 


Comments

bottom of page