Rocket.new, one of India’s first vibe-coding startups, snags $15M from Accel, Salesforce Ventures

Rocket.new, an Indian startup gathering an AI-powered app improvement platform, has raised $15 cardinal successful a effect circular led by Salesforce Ventures to instrumentality connected viral vibe-coding rivals similar Lovable, Cursor, and Bolt by letting users physique full, production-ready apps from natural-language prompts alternatively than conscionable speedy prototypes.

Accel and Together Fund joined Salesforce Ventures successful the all-equity effect round, which comes lone 3 months aft Rocket.new launched its level successful beta successful June.

Since its debut, the startup has crossed 400,000 users — including implicit 10,000 paid subscribers — successful 180 countries and reached $4.5 cardinal successful yearly recurring revenue. It is aiming to standard that to $20–$25 cardinal by year’s extremity and $60–$70 cardinal by June adjacent year, co-founder and CEO Vishal Virani (pictured above, center) said successful an exclusive interview.

Based successful Surat — a metropolis celebrated for diamonds and textiles but acold from India’s accustomed tech hubs — Virani teamed up with Rahul Shingala and Deepak Dhanak to motorboat Rocket.new. The startup marks a pivot from their earlier venture, DhiWise, which focused connected developer workflows.

“We are gathering the archetypal vibe solution platform, which is not solving conscionable a occupation of time one, but what we are focusing connected is solving the occupation of time two,” Virani told TechCrunch.

The 16-week-old startup aims to go a broad agentic system, which volition utilize AI not lone to physique apps and websites but besides to behaviour competitory probe and merchandise development, eliminating the request for merchandise managers.

“Our full agentic strategy volition assistance organizations physique each kinds of functions astir products — not conscionable generating the root codification — but adjacent a installation to standard their merchandise — each by giving natural-language prompts,” Virani asserted.

The existent model, which Virani refers to arsenic mentation 0.3, has already built fractional a cardinal applications and attracted merchandise managers, solopreneurs, and front-end developers. It besides counts users from companies similar Meta, PayPal, KPMG, PwC, and Times Internet, who crook to the level for idiosyncratic projects.

Image Credits:Rocket.new

About 80% of Rocket.new’s users person truthful acold built what Virani calls “serious” applications alternatively than elemental landing pages oregon branding sites. Roughly 12% person created e-commerce platforms successful segments similar market and apparel, 10% person built fintech apps, 5–6% person developed B2B tools, and 4–5% person launched intelligence wellness apps, helium said.

Roughly 45% of Rocket.new’s users are gathering mobile apps and 55% are gathering websites, Virani said. He added that galore users physique a website connected Lovable oregon Replit, past crook to Rocket.new to make a autochthonal mobile app by integrating with their existing Supabase backend.

The level combines ample connection models from Anthropic, OpenAI, and Google’s Gemini with Rocket.new’s ain heavy learning systems, which are trained connected proprietary datasets from DhiWise.

“Our underlying architecture is wholly antithetic from what Lovable, Bolt, and everyone is doing,” Virani told TechCrunch.

Generating the archetypal exertion takes astir 25 minutes — overmuch slower than astir vibe-coding tools, which usually nutrient results wrong three. But successful our aboriginal testing, Rocket.new delivered a much broad idiosyncratic experience, with each indispensable modules included.

Rocket.new offers a escaped proceedings capped astatine 1 cardinal tokens. After that, users wage a monthly subscription starting astatine $25 for 5 cardinal tokens.

This exemplary efficaciously discourages hobbyists, portion giving Rocket.new a steadfast gross borderline of 50–55% — a fig it aims to rise to 60–70% wrong the adjacent fewer months.

The U.S. is Rocket.new’s largest market, contributing 26% of revenue, followed by Europe astatine 15–20% and India astatine 10%. To amended service its American users, the startup is establishing a U.S. office successful Palo Alto.

Virani noted that Rocket.new’s aboriginal traction was organic, driven by word-of-mouth and viral societal posts. With caller effect funding, the startup is looking to sharpen its go-to-market playbook, deepen its beingness successful cardinal markets, and accelerate enactment connected proprietary models and R&D.

“We saw a wide spread betwixt the magic of AI codification procreation and the world of making that codification production-ready,” Kartik Gupta, Investor astatine Salesforce Ventures, told TechCrunch implicit email. “Rocket.new is purpose-built to lick this occupation of iteration, maintenance, and deployment astatine endeavor scale.”

Rocket.new presently has a 58-member team, chiefly successful Surat, and expects to treble its engineering and merchandise unit successful India implicit the adjacent 12 months to enactment growth.