For years, the story of AI in Indian e-commerce has mostly been told through customer-facing chatbots and recommendation engines. Myntra's latest technology showcase suggests the more consequential transformation may actually be happening on the other side of the platform entirely — in how fast a seller can get a new brand live, how quickly a product catalogue can be generated, and how rapidly internal engineering teams can ship new features.
The Showcase: mynnovAIte
The Walmart-owned fashion e-commerce platform laid out its AI strategy at mynnovAIte, a technology showcase held at its Bengaluru headquarters on July 16, 2026. Led by Chief Product Officer Lakshmi Narayan Swaminathan and Chief Technology Officer Pramod Addidam, the event focused on how artificial intelligence is being deployed across three distinct pillars of the business: sellers, customers and internal operations. The framing offered by Myntra's leadership was notably understated for a technology showcase of this scale — Swaminathan characterised the effort not as AI being used to solve entirely new problems, but rather as a tool for solving familiar, long-standing operational bottlenecks dramatically faster and more efficiently than before.
That framing matters, because it positions Myntra's AI rollout less as a speculative bet on frontier technology and more as a systematic re-engineering of processes that have constrained the platform's growth for years — chief among them, how long it takes a new seller or brand to actually start selling on the platform.
Seller Enablement: The Headline Number
The single most striking figure to emerge from mynnovAIte concerns seller onboarding. Myntra has reduced the time it takes for a new seller to go live on its platform from 10 to 15 days down to approximately one to two days — a reduction of roughly 85 to 90 percent in onboarding time. For a marketplace platform where seller acquisition and time-to-revenue are directly tied to growth, compressing this window from over a week down to a matter of days represents a genuinely significant operational unlock, not merely an incremental efficiency gain.
Swaminathan indicated that seller feedback directly shaped many of the changes underlying this improvement: sellers had consistently asked for simpler ways to resolve incomplete or stuck listings, and clearer, more actionable guidance on what specific steps to take next, rather than being left to interpret large volumes of raw performance data on their own. In response, Myntra built Saarthi, a voice-based AI workflow tool that — notably — has been built as a configurable platform that other internal teams beyond the seller-facing function can adapt to create their own voice-based workflows, suggesting the underlying architecture was designed with reusability across the organisation in mind, rather than as a narrow, single-purpose tool.
Importantly, when asked whether these AI-powered seller tools would eventually be walled off behind paid tiers, Swaminathan was direct: Myntra currently has no plans to differentiate access based on seller tiers, framing the tools as designed to improve overall marketplace performance rather than serve as a premium, monetisable offering. In a marketplace ecosystem where platform fees and paywalled features are common points of seller friction, this stance — at least as currently articulated — positions the AI rollout as a genuine platform-wide capability improvement rather than a new revenue lever layered on top of existing seller relationships.
Catalogue and Content: From a Day to Four Hours
On the content and catalogue side, Myntra has compressed the time required to generate a product catalogue from roughly one full day down to about four hours — a reduction that becomes especially significant when multiplied across the thousands of new product listings the platform processes on an ongoing basis. The platform now generates between 400 and 600 dynamic product videos daily, an output volume that would be effectively impossible to sustain through manual video production at Myntra's scale, and that speaks to how deeply AI-generated content has been woven into the platform's day-to-day content operations rather than being deployed for a narrow subset of premium listings.
The company has also strengthened its ability to associate detailed product attributes automatically — reportedly linking up to 40 distinct attributes per product — improving the granularity of data available for search, filtering and recommendation systems, all of which depend on rich, accurate product metadata to function well.
Customer Experience: Solving Fashion's Fit Problem
On the customer-facing side, Myntra has introduced an AI-powered "Size and Fit Intelligence" layer that now covers approximately 85 percent of its eligible apparel catalogue — directly targeting one of online fashion retail's most persistent and costly problems: the uncertainty shoppers face about whether a garment will actually fit before receiving it. Size and fit uncertainty is widely understood across the fashion e-commerce industry as one of the leading drivers of returns, and a meaningful improvement in fit prediction accuracy carries direct implications not just for customer satisfaction, but for the underlying unit economics of the business, given how costly reverse logistics and returns processing can be at scale.
Beyond fit, personalisation has become deeply embedded across the platform: Myntra reports that 90 percent of its monthly active users now experience personalised search recommendations, a figure that suggests personalisation has moved from being a differentiating feature for a subset of engaged users to something approaching a default, platform-wide experience. The company is also expanding its AI-led Contextual Styling Suite, which offers real-time look pairings across the catalogue — helping shoppers move beyond simply searching for individual products toward understanding how those products might be styled together as a complete outfit, and, as the company indicated, eventually toward guidance on how and where a given look could actually be worn.
Chief Technology Officer Pramod Addidam framed the underlying philosophy behind this styling-focused investment in terms specific to fashion as a category: fashion commerce, in his framing, poses distinct challenges beyond typical product search, given how quickly trends evolve and how deeply personal individual style preferences are. The company's ambition, as articulated at mynnovAIte, is to build toward what it describes as a fashion-native conversational experience — one where users can discover products, styling ideas and complete looks through natural interaction, rather than simply receiving text-based search results in response to narrow queries. This vision builds on Maya, Myntra's existing conversational AI assistant, extending its scope from basic product search toward a considerably more ambitious styling and discovery experience.
Internal Operations: The Least Visible, Most Structural Change
Perhaps the least publicly visible but most structurally significant category of AI deployment discussed at mynnovAIte concerns Myntra's internal engineering and operations functions. The company reported that AI has boosted the speed of feature rollouts by 40 percent, meaning the platform's own engineering teams are now able to ship new product features considerably faster than before — a compounding advantage, since faster feature velocity means faster iteration on every other AI system described above.
In supply chain operations, AI has reportedly shortened the time required for complex network simulations from two full days down to a single hour — a dramatic compression that has direct implications for how quickly Myntra's logistics and inventory-planning teams can model and respond to changing demand patterns, seasonal spikes, or supply disruptions, rather than working from simulations that were, until recently, too slow to support truly responsive planning.
Governance and Guardrails
Notably, Myntra's public communication around this AI rollout has consistently emphasised that all of its AI deployments continue to operate with human oversight, privacy safeguards and governance built into relevant workflows, despite the increasing degree of automation across seller, customer and internal-operations functions. This framing — emphasising human oversight even as automation accelerates — reflects a broader pattern increasingly common among large consumer platforms deploying AI at scale: a recognition that speed and efficiency gains need to be paired with visible governance commitments, particularly for systems that touch customer-facing decisions like size recommendations or seller-facing decisions like listing approvals, where errors carry direct commercial and reputational consequences.
What This Signals for Indian E-Commerce More Broadly
Myntra's mynnovAIte showcase arrives at a moment when AI deployment claims have become almost a prerequisite talking point for any major Indian consumer internet platform, making it worth asking what, if anything, distinguishes this particular rollout from the broader wave of AI announcements across Indian retail and e-commerce. The answer appears to lie less in any single novel technology and more in the breadth and specificity of the measurable operational improvements Myntra is willing to disclose: an 85-90 percent reduction in seller onboarding time, a roughly 80 percent reduction in catalogue generation time, and a 40 percent improvement in feature-rollout velocity are all concrete, auditable metrics rather than vague claims about "AI transformation."
For a platform navigating an increasingly competitive Indian fashion e-commerce landscape — one that now includes quick-commerce players expanding into fashion and lifestyle categories, alongside established rivals — the ability to onboard new brand partners in days rather than weeks, and to generate rich, personalised product content at a pace that scales with catalogue growth rather than being constrained by manual content production, represents a genuine structural advantage. Whether that advantage translates into sustained market share gains over the coming year will depend on execution well beyond the showcase stage — but the specificity of the metrics Myntra chose to disclose at mynnovAIte suggests a company confident enough in its underlying numbers to invite that scrutiny.
The Seller-Side Story Deserves More Attention Than It Usually Gets
Much of the public conversation around AI in Indian e-commerce tends to concentrate on the customer-facing layer — chatbots, recommendation engines, virtual try-ons — largely because those are the applications an ordinary shopper directly experiences. Myntra's seller-onboarding numbers deserve considerably more attention than they typically receive, precisely because seller-side friction has historically been one of the most persistent, least glamorous bottlenecks constraining marketplace growth across Indian e-commerce generally, not just at Myntra specifically.
A seller onboarding window of 10 to 15 days represents a genuinely significant barrier to marketplace liquidity: every additional day a prospective brand partner spends navigating documentation, catalogue setup, compliance checks and listing approval is a day that brand is not generating revenue for either itself or the platform, and a day during which a smaller or less patient brand might simply abandon the process altogether in favour of a competing marketplace with a faster path to going live. Compressing that window to one to two days does not merely improve an internal operational metric — it plausibly expands the total pool of brands willing to attempt onboarding at all, particularly smaller and newer D2C labels that may lack the dedicated operations staff to navigate a two-week onboarding process, but that can readily manage a one-to-two-day one.
Where the AI Rollout Could Still Face Real-World Friction
Ambitious internal metrics announced at a company-run technology showcase inevitably invite a degree of healthy scepticism, and Myntra's mynnovAIte disclosures are no exception. Onboarding-time reductions and catalogue-generation speedups are relatively straightforward to measure and report accurately, but the downstream quality implications of AI-generated content — product descriptions, size and fit predictions, styling recommendations — are considerably harder to verify from the outside, and will likely only become clear over the coming quarters as customer-facing metrics like return rates, customer satisfaction scores and seller retention rates are reported, if Myntra chooses to disclose them with the same specificity it applied to its process-efficiency claims.
There is also a broader industry question the mynnovAIte showcase does not fully resolve: as AI-generated product videos, AI-assisted seller onboarding and AI-driven styling recommendations become standard features across major e-commerce platforms rather than differentiators unique to any single company, how much of a durable competitive advantage can any one platform's specific AI implementation actually sustain over time, versus simply keeping pace with an industry-wide baseline that rival platforms are racing to match in parallel. Myntra's early-mover specificity in disclosing hard operational metrics may provide a temporary communications advantage, but the underlying AI capabilities it described — automated content generation, personalised search, fit prediction — are being pursued in parallel by competing platforms across Indian and global e-commerce, meaning today's differentiator could plausibly become tomorrow's table stakes.
