Marketing Personalization
Fisia needed to improve fit confidence, conversion, and basket size for Nike shoppers while reducing returns and making the same personalized experience work consistently in stores, on mobile devices, and online.
Reported impact
Fisia raised in-store conversion by 150 percent, increased average ticket size by 45 percent, achieved 128 percent ROI, and used the same real-time recommendation stack to improve fit confidence and reduce friction across channels.
Fisia deployed Doris.ai as an experience-commerce layer for Nike Brazil so shoppers could move between physical and digital channels without losing personalization. The workflow starts with two smartphone photos that Doris.ai uses to estimate body measurements and generate a personalized 3D avatar. Behind that experience, the platform runs seven APIs on a Kubernetes cluster backed by NVIDIA A10 Tensor Core GPUs, while NVIDIA TensorRT and Triton Inference Server optimize the recommendation and avatar-inference path for real-time performance. Product-catalog data, purchase history, and live inventory are combined to recommend sizes, outfits, alternate colors, and complementary items in one conversational flow. This gives Fisia a scalable try-on and sizing stack that can serve smart mirrors, mobile apps, and ecommerce surfaces from the same backend.
Two shopper photos are processed by Doris.ai to estimate measurements and generate a personalized 3D avatar. Product catalog, purchase history, and real-time inventory data then flow through TensorRT- and Triton-optimized services on A10 GPU-backed Kubernetes clusters to return size recommendations, outfit suggestions, and virtual try-on results across store, web, and mobile channels.
Behind the scenes, Doris.ai leverages NVIDIA TensorRT and Triton Inference Server to deliver tailored recommendations and answer nuanced customer queries.
In a recent proof of concept with Fisia Group, in-store Doris.ai users saw conversion rates rise from 8% to 20%, while average ticket size climbed by 45%.
The platform's virtual try-on and avatar generation capabilities are powered by a combination of seven APIs, hosted on a Kubernetes cluster of AWS instances equipped with NVIDIA A10 Tensor Core GPUs.
With just two photos from a shopper's phone-one from the front and one from the side-the platform instantly captures accurate body measurements.
Review how this reference was scored and trace each normalized claim back to its documented source.
Scoring methodology
Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.
Traceability
Field coverage shows which facts each source supports. Dates are shown only when documented by the source.
Related
NVIDIA Sample Aug 14, 2026 Shopee Shopee needed large language models that could handle multilingual ecommerce tasks across Southeast Asia, Taiwan, and Brazil, but general-purpose frontier models degraded sharply on regional languages, shopping… 113x growth from 3 billion to 340 billion monthly tokens Compass v3.5 · NVIDIA A100 GPUs · NVIDIA H100 GPUs NVIDIA Sample Jul 24, 2026 Instacart Instacart needed real-time visibility into in-store shopping behavior, shelf conditions, and basket changes so retailers could reduce shrink, improve recommendations, and personalize physical-store experiences that… 3x year-over-year Caper Cart growth NVIDIA Dynamo · NVIDIA Hopper GPUs · NVIDIA Jetson Orin NX modules 
Alibaba Cloud Sample Oct 13, 2026 TOPSPORTS (滔搏) TOPSPORTS wanted to scale enterprise AI agents beyond isolated pilots, but every new agent risked recreating the same knowledge, data, tools, and skills in separate silos. Business-analysis, membership, and marketing… business analysis time reduced to 20 minutes AgentCore · AgentLoop · MCP tools 