Prepared for Swiggy · operations

Swiggy is winning the quick-commerce land grab and paying for it by the order.

Every point of margin now decides who reaches profit first.

The order is growing faster than the economics

In Q4 FY26, Instamart's gross order value jumped 68.8% year-on-year to ₹7,881 crore¹. In the same quarter its contribution margin sat at -1.8% of GOV — improving to -1.1% for March, but negative all the same¹. Adjusted EBITDA held at -10.9% and the consolidated loss narrowed to ₹800 crore², with the profitable food-delivery engine — ₹272 crore of adjusted EBITDA, up 13.1% sequentially⁴ — funding the growth. With rivals already at quick-commerce profitability, the gap to Instamart's own 4-5% medium-term target³ is now an operations race, not a capital one.

Where applied AI earns its keep

Closing that gap is an operations problem before it is a capital one — which is exactly where a focused build pays back. Meridian AI is a boutique applied-AI studio that embeds with your team and ships production-grade LLM and agent systems on your existing stack: retrieval over your own order and rider data, function-calling agents wired into the services Instamart already runs, evals and guardrails in CI. No model training, no rip-and-replace, no new infrastructure to run, and a human-review path on every decision that touches a customer.

Tied to one number, not a promise

Every engagement targets one measurable operating metric per quarter — cost-to-serve per order, deflected support contacts, rider idle time — and we don't scale the retainer until that metric moves. You see the lever work before you fund it further.