OPINION | 2025 Roundup: The year Indian startups got its mojo back

The Year of AI Applications

While the intelligentsia debates India’s AI future, the builders have quickly figured out what to build. India has a long history of building enterprise applications for the world, first as IT services and, in the past decade, as SaaS applications. There is a once-in-a-generation opportunity to rebuild the entire enterprise software stack using AI, and Indian founders are going after it. In 2025, we backed AI applications in areas ranging from insurance underwriting to software testing to personal injury law. Unlike traditional software that only drives automation, AI can augment or replace human judgment, allowing businesses to make decisions better and faster. This capability is invaluable for “white-collar” businesses; the erstwhile knowledge economy is now the AI economy, and Indian VCs are investing aggressively into it.

Giving Voice to Intelligence

Within the overall AI applications space, one opportunity stands out: voice applications. India is a voice-first market: most services are sold, not bought; most customers need hand-holding during the buying process; companies spend enormous budgets on telesales. Voice AI is a game-changer in such a market: it offers infinite scalability, immediate lead callback, and likely a lower cost of sales than human agents. In 2025, we saw voice AI applications allow an NBFC to grow lead coverage by 10x; a travel company increase sales productivity by 30%; and an edtech company grow lead quality by 6x. When we heard a customer trying to flirt with the AI voice agent, we knew this was a technology whose time had come.

The India Advantage in AI

There’s been a lot of chatter about India’s “right to win” in AI applications. The initial answer seems to lie where it often does in India: in our human capital. Due to accuracy limitations, the criticality of certain business processes, and regulatory constraints, many AI applications require a “human in the loop.” In some cases, the customer provides such human oversight (to a code-writing bot, for example); but in many others, the vendor is expected to deliver the final output using a combination of AI and human judgment. A typical Silicon Valley startup is often not geared to provide such “full-service” applications – they not only face a labour shortage but are also culturally and managerially unused to managing a large white-collar workforce. Indian startups, on the other hand, think nothing of hiring 200 data analysts in a single year to augment the work of the AI.

In 2025, we have backed such full-service AI applications in sectors ranging from accounting to customer support to enterprise sales. Compared to traditional services businesses, these companies are seeing 2-4x improvements in gross margins by delivering the same business processes. No wonder AI-led services have been the defining theme of Indian VC in 2025.

The Steady March of Quick Commerce

The single biggest driver of our consumption patterns in the past five years has been the rise of quick commerce. What started as 10-minute groceries has evolved into 10-minute everything. Indian VCs are betting that many quick commerce categories will be better served by dedicated retailers than by horizontal quick commerce players – their supply chains, demand patterns, store operations, and delivery logistics will require dedicated focus. Vertical startups in this sector have ranged from those providing fashion to baby care products.

Additionally, there is an opportunity to disrupt the services business with quick commerce. This year, Indian VCs funded startups providing domestic help, home maintenance, and beauty services in 10 minutes, and we have started to see strong adoption in their initial target markets.

2026 Predictions

In 2026, we are likely to see an acceleration of some investment trends from this year and a maturing of others. Vertical commerce investments might see a slowdown as leaders begin to emerge in most verticals. Investments in AI applications, on the other hand, might go more vertical as entrepreneurs delve deeper into automating stubbornly human-led enterprise processes. AI applications are also likely to move from being point tools to workflow agents – instead of automating one pre-defined process, these applications will be able to map multi-step enterprise processes and perform them with little human assistance.

2026 will also see Indian entrepreneurs start global businesses from day one: the product and GTM playbook for such startups is starting to emerge, and we expect the trend to pick up speed next year. We also expect younger founders to get funded, particularly in enterprise-focused startups that had earlier been the domain of more experienced founders. When the playbook needs to be reinvented rather than followed, the young and fearless often have an advantage. And us investors are here for it.

Amen!

(Ritesh Banglani, Partner at Stellaris Venture Partners.)

Views are personal, and do not represent the stance of this publication.