Meesho Leverages AI to Drive Efficiency with Over 70% of Codebase Now Automatically Generated

Meesho Embeds AI Across Operations, Aims to Accelerate Product Development and E-commerce Growth

Meesho, the SoftBank-backed e-commerce platform, is rapidly integrating artificial intelligence (AI) across its operations, with more than 70 percent of the company's code now being AI-generated. This move is a deliberate bet on AI as the operating system for how Meesho builds products, according to chief executive Vidit Aatrey.

As part of its Q4 FY26 results, Meesho released its annual shareholder letter, which highlights the company's significant investments in AI-led commerce infrastructure. AI has been integrated across the software development lifecycle, including code and test-case generation, code reviews, deployment fixes, and production monitoring. This integration has led to a significant reduction in the time it takes to develop and release products, with Meesho now releasing products faster and with greater reliability than at any point in its history.

The company's focus on AI is not just about cost reduction, but rather about accelerating product development and solving infrastructure problems unique to India's e-commerce market. Meesho's chief financial officer, Dhiresh Bansal, stated that the company is not looking at AI as a cost lever, but rather as a tool to improve efficiency and expand e-commerce access across India.

Read also: Google Pixel Watch Leak: Diver's Discovery Sparks Speculation

Meesho's AI Push: Key Initiatives and Results

Meesho's AI push is also centered around conversational and voice-led commerce, which the company believes could become the dominant shopping interface for the next wave of internet users in India. The company's voice AI shopping assistant, Vaani, has seen strong early traction, with users who adopted the feature seeing a 22 percent improvement in conversion rates.

Beyond shopping interfaces, Meesho is using AI to solve operational challenges tied to India's fragmented logistics and commerce ecosystem. The company's proprietary AI model, GeoIndia LLM, has improved geocoding accuracy by 20 percentage points and reduced misroute-related costs by 5 percent. Meesho is also deploying AI models across fraud detection, logistics planning, and customer support.

Read also: Chinese Startup Claims AI Coding Capabilities Exceed Those of GPT-5.5: A Detailed Examination

AI Becoming Core to Meesho's Operating Model

Meesho's management sees AI as central to how the company builds products, improves efficiency, and expands e-commerce access across India. The company is investing in what it calls an "autonomous SDLC", where AI agents increasingly handle software development workflows, including coding, testing, review, and deployment. Meesho is also positioning AI not just as an efficiency tool, but as the core infrastructure layer powering the next phase of Indian e-commerce growth.