Ahmet Kayıran on "Korporasyon"

In this episode of Korporasyon, RNV.ai Co-founder and Chief Data Officer Ahmet Kayıran discusses how artificial intelligence is transforming traditional retail challenges into modern, data-driven solutions. Sponsored by RNV.ai, the conversation explores how delivering "the right product, at the right time, at the right price" is no longer just about experience — it requires powerful data engineering.

From Retail Analytics to AI-Powered Solutions

Ahmet Kayıran shares his journey from industrial engineering to building data-driven decision systems in the retail sector. With a background in retail analytics, he co-founded RNV Analytics alongside Gökhan Yücel to tackle the industry's most persistent challenge: getting the right product to the right store, at the right price.

The founding partners realized that their work on inventory optimization processes could be productized, brought to market, and scaled broadly. This insight became the foundation of RNV.ai and its SaaS product, Klibre.

Why AI in Retail? Processing 1 Million SKUs in 15 Minutes

Retail's fundamental problem is familiar: stock exists, but not in the right place, at the right time, or at the right price. RNV.ai's algorithms analyze data at a scale that humans could never review in weeks — processing it every night in just 15–20 minutes.

The system evaluates all scenarios and calculates which product should be in which store, in which size breakdown, in what quantity, and at what price. These decisions are then automatically converted into actionable work orders.

• Inventory optimization across the entire retail network
• Pricing and promotion planning with AI-driven accuracy
• Supply chain and store-to-store transfer optimization
• Demand forecasting and commercial planning
• E-commerce process improvements

Conventional AI vs. Generative AI in Retail

Kayıran draws an important distinction between conventional AI (machine learning for prediction, modeling, and optimization) and generative AI. RNV.ai primarily leverages conventional AI with ML algorithms that provide evidence-based solutions from large datasets.

The platform uses composite methods — ensemble models and hybrid approaches that first make predictions and then perform optimization based on those predictions. These models work together harmoniously to orchestrate complex decision points across the retail value chain.

Feature Engineering: The Hidden Key to AI Success

One of the most technical and insightful parts of the conversation dives into feature engineering — the critical process of extracting meaningful attributes from noisy retail data.

Even in our simplest model, there are nearly 300 intrinsic attributes, and each of them can create correlations with each other, or even multiplier effects. Product color, fabric type, technology variants, hierarchy, pricing — all become inputs that determine model effectiveness.

Explainability and Transparency

A key differentiator for RNV.ai is its commitment to explainable AI. The platform doesn't just make predictions — it explains why decisions were made. Is it related to pricing policy? Similar products? Store segmentation? Capacity constraints?

This transparency model answers all these questions and reports them to decision-makers, enabling continuous improvement and building trust in AI-driven recommendations.

Career Advice for Data Science Professionals

In the final segment, Kayıran shares valuable advice for young professionals aiming for careers in data science and AI. He emphasizes that understanding the domain — in this case retail — is just as important as knowing the algorithms. Feature engineering, he notes, requires deep process knowledge that goes beyond technical model expertise.

This article is a summary of the episode. Watch the full conversation to explore the founding story of RNV.ai, how AI is solving traditional retail problems, and the future of the sector from a logistics and engineering perspective.