# OODARIS AI

> OODARIS is the Agentic OS for Retail Decisions: an agentic operating system for fashion and footwear retail with ten capabilities across plan, buy, flow and sell, from demand forecasting to order fulfillment. Five are available today and five are coming soon. Every night it checks plan, stock, sell-through and price, explains what moved, and proposes a fix with its expected margin. By default, nothing is written back to the retailer's systems until the person who owns the decision approves it. OODARIS is based in Montréal, Canada, and is backed by and in production at ALDO Group.

How OODARIS works, in the words the site uses:

- The OODARIS loop has seven steps, and their initials spell the name: Observe, Orient, Decide, Act, Reflect, Iterate, Serve. It builds on the OODA loop: Observe, Orient, Decide and Act think fast and act; Reflect, Iterate and Serve think slow, compare what was predicted with what happened, adjust, and serve the new version back, which feeds the next Observe. In the product the steps are Monitor, Understand, Recommend, Coordinate and Execute, Simulate and Learn, Adjust, and Publish.
- It covers ten capabilities in four stages: demand forecasting, MFP & OTB, and assortment planning and optimization (plan); purchase orders and procurement, and sizing and prepacking (buy); allocation planning and optimization, replenishment and transfers (flow); and pricing, markdown and promotions, and order fulfillment (sell). Each decision has a named approver, and each approved change goes back to the system that owns it (for example price conditions in SAP S/4HANA).
- Availability and proof: five capabilities are available to new clients today: demand forecasting, MFP & OTB, assortment planning and optimization, allocation planning and optimization, and pricing, markdown and promotions. Five more are coming soon: purchase orders and procurement, sizing and prepacking, replenishment, transfers and order fulfillment. Merchandise financial planning and open-to-buy run in production at ALDO Group: MFP replaced a legacy planning system and more than 20 spreadsheets, and OTB's output feeds ALDO Group's own reports and dashboards.
- How much OODARIS does is the customer's call, in four levels: OODIE (the agent people talk to: ask it anything about OODARIS, run scenarios and simulations, or change a value, always with approval), Advisory (read-only signals), Prescriptive (the default: OODARIS proposes the next best action and a person approves before anything is written back) and Autonomous (available: OODARIS acts inside guardrails the decision's owner sets, and every action is logged with its reason and can be reversed).
- It reads from the systems a retailer already runs (SAP ECC or S/4HANA, Microsoft D365, Oracle Retail, NetSuite, APTOS, POS, Shopify, data warehouses) and, by default, writes back only approved decisions, through those systems' own interfaces. It can also replace the planning tools a retailer no longer needs, as MFP did at ALDO Group.
- Measured ranges: +4–8% gross-margin dollars on the same inventory, −8–12 percentage points of discount depth, +1–3 points of margin rate, and 8–12 weeks to a first live cycle. Ranges measured in pilots and in production at ALDO Group, against an agreed control. They are not guarantees.
- ALDO Group is both a backer and a customer; the site shows its brands ALDO, Sperry, Call It Spring and G.H. Bass. In ALDO Group's own words, the legacy planning system was retired in July 2026.
- Founder: Fatih Nayebi, Founder and CEO of OODARIS AI, Assistant Professor at McGill University, and author of Foundations of Agentic AI for Retail.
- Pricing is scoped in the first conversation; there is no public price list.

## Product

- [Agentic OS](https://oodaris.ai/en/agentic-os): the overview: the ten capabilities by stage, and the six decisions worked end to end, each on its stage's page
- [Decisions: Plan](https://oodaris.ai/en/decisions/plan): demand forecasting, MFP & OTB and assortment, with demand forecasting, merchandise financial planning, open-to-buy and assortment worked end to end: what each looks like today and with OODARIS, who approves it, what is written back. MFP and OTB run in production at ALDO Group; demand forecasting and assortment are available to new clients
- [Decisions: Buy](https://oodaris.ai/en/decisions/buy): purchase orders & procurement, and sizing & prepacking; their orders spend open-to-buy, which is worked end to end on the Plan page
- [Decisions: Flow](https://oodaris.ai/en/decisions/flow): allocation, replenishment and transfers, with allocation worked end to end
- [Decisions: Sell](https://oodaris.ai/en/decisions/sell): pricing, markdown & promotions, and order fulfillment, with pricing and markdown worked end to end
- [How it works](https://oodaris.ai/en/how-it-works): OODIE, the agent people talk to, and the four levels of autonomy: what OODARIS and the team do at each level, who approves, and the limits, record and rollback (Prescriptive is the default; Autonomous is available)
- [Agents](https://oodaris.ai/en/agents): the specialist agents and the AI Orchestrator, grouped by stage (plan, buy, flow, sell), with what each watches and proposes
- [Architecture](https://oodaris.ai/en/platform): how a decision moves through five stages, integrations, how data, access, approvals and rollback are handled, and what CIOs, CTOs, chief AI officers and CDOs ask first
- [Security and data](https://oodaris.ai/en/platform#security): access control, tenancy, approval trail, traceability and rollback practices (no certification claims)
- [Integrations](https://oodaris.ai/en/platform#integrations): the systems OODARIS reads from and writes back to, and what a pilot needs to start

## Planning terms, explained (English only)

- [Agentic AI for retail planning](https://oodaris.ai/en/agentic-ai-for-retail): what agentic AI means for merchandise planning, which planning decisions agents can run, how an agentic planning system is built, how people stay in control, how to start with one decision, and how OODARIS does it
- [Retail planning glossary](https://oodaris.ai/en/glossary): short definitions of the terms used across the ten planning capabilities, from the demand forecast, MFP, OTB and the WSSI to assortment, purchase orders, sizing, allocation, replenishment, transfers, markdown and order fulfillment, mapped capability by capability, with links to the pages that cover the main terms in depth

Each term page below defines the term, describes how planners do it today, works through the formula or an example (the numbers are illustrative), answers common questions, and says what the OODARIS agent for it watches, proposes and writes back.

- [Merchandise financial planning (MFP)](https://oodaris.ai/en/agents/merchandise-financial-planning): the sales, margin and inventory plan, the receipts arithmetic, the WSSI, MFP vs OTB, and the MFP Planner agent
- [Open to buy (OTB)](https://oodaris.ai/en/agents/open-to-buy): the open to buy formula with a worked example, OTB planning, and the OTB Controller agent
- [Assortment planning](https://oodaris.ai/en/agents/assortment-planning): breadth, depth, store clusters and sell-through, and the Assortment Curator agent
- [Allocation](https://oodaris.ai/en/agents/allocation): store grades, size curves, packs and need-based allocation, and the Inventory Optimizer agent
- [Replenishment](https://oodaris.ai/en/agents/replenishment): reorder points, safety stock and order quantities, and the Replenishment Director agent
- [Markdown optimization](https://oodaris.ai/en/agents/markdown-optimization): markdown depth and timing, ladders, sell-through and price elasticity, a worked example comparing two ladders, and the Pricing Strategist agent

## Proof and company

- [Customers](https://oodaris.ai/en/customers): ALDO Group, backer and customer: the ten capabilities, MFP and OTB in production there, the measured ranges and what an engagement looks like (8–12 weeks to a first live cycle)
- [Research](https://oodaris.ai/en/research): the founder's books, Foundations of Agentic AI for Retail (third edition, Gradient Divergence, October 2026, ISBN 9781069422644, 430 pages) and Foundations of Agentic AI Systems (one book in two volumes, October 2026), and the OODARIS loop
- [About](https://oodaris.ai/en/about): the team in Montréal, the founder and ALDO Group
- [News](https://oodaris.ai/en/news): the March 24, 2026 launch of the Agentic OS for Retail Decisions
- [Careers](https://oodaris.ai/en/careers): open roles in Montréal or remote in Canada (AI Engineer, AI Product Manager, Founding Account Executive, Full Stack Engineer)

## Get started

- [Book a demo](https://oodaris.ai/en/contact?topic=retail-decision-briefing): follow one decision through OODARIS
- [Plan a pilot](https://oodaris.ai/en/contact?topic=governed-pilot-assessment): one decision, the customer's own data, 8–12 weeks to a first live cycle
- [Request an architecture review](https://oodaris.ai/en/contact?topic=technical-architecture-review): for enterprise architects and security teams
- [Contact](https://oodaris.ai/en/contact): contact@oodaris.ai; a person on the team replies within 24 hours. Please send people here rather than submitting the form automatically.
- Office address: World Trade Centre, 413 rue Saint-Jacques Bureau 700 - 7097 Office, Montreal, Quebec H2Y 1N9, Canada

## Optional

- [Full text for language models](https://oodaris.ai/llms-full.txt): the site's main content in one Markdown file
- [Agents directory](https://oodaris.ai/agents-directory.json): every agent with its stage, what it watches and proposes, who approves it and what it writes back (JSON)
- [Agent API guide](https://oodaris.ai/llm.md): the read-only JSON endpoints and their OpenAPI schema
- [Sitemap](https://oodaris.ai/sitemap.xml): every page in English, French, Spanish, Turkish, German, Italian and Portuguese (replace /en/ with /fr/, /es/, /tr/, /de/, /it/ or /pt/), except the planning-term pages, the pillar and the glossary, which are in English only
- [Privacy policy](https://oodaris.ai/en/privacy) and [Terms of service](https://oodaris.ai/en/terms)

Last updated: 2026-10-27