# ZAAI

> ZAAI is an applied AI engineering company that builds and deploys enterprise AI systems. Founded in 2023, based in Porto, Portugal. Deployed across 11+ industries.

## Products

### Horizon
Forecasting and decision intelligence engine. Connects to ERP systems (SAP, Oracle, NetSuite), e-commerce platforms (Shopify), and banking systems. Ingests historical transaction data, learns demand patterns at configurable granularity (SKU x customer x week, product x channel x day). Produces: (1) demand forecasts with 85-92% accuracy across 16-week horizons, (2) scenario simulations for pricing, promotions, supply disruptions, and demand shocks, (3) automated action recommendations for replenishment, production scheduling, and inventory rebalancing. An agent layer executes routine decisions (reorders, inter-store transfers, schedule updates) within team-set rules (budget thresholds, quantity limits, approval flows), with escalation to a person and a full audit trail. Uses hierarchical forecast reconciliation with weighted least-squares. Deployed for manufacturers managing 2,000+ concurrent time series and retailers with 1,700+ SKUs across 42 stores. Horizon is not a general-purpose analytics platform, a BI dashboard, or a data warehouse. It is a specialized forecasting engine that decides and acts.

### Pliney
Agentic AI system for customer-facing operations. Deploys purpose-built AI agents into existing chat interfaces (web, mobile, messaging platforms) grounded in real-time operational data. Four agent types: (1) Customer Support Agent: resolves order tracking, returns, billing queries by accessing order history, account data, and company policies, (2) Sales Agent: qualifies leads, answers product questions, books meetings with calendar integration, (3) Account Manager Agent: handles subscriptions, settings, usage reporting for authenticated users, (4) Product Navigator: natural language product search, comparison, and personalized recommendations. Integrates with Salesforce, HubSpot, Zendesk, Intercom, Shopify, SAP, Oracle, and custom APIs. Security: end-to-end AES-256 encryption, per-customer data isolation, no PII retention, SOC 2 Type II-aligned controls. Pliney is not a rules-based chatbot, a FAQ widget, or a search bar. It is an agentic system that reasons over live data and takes actions.

### Engineering Engagements
Custom enterprise AI solutions for problems that do not fit standard products. Examples: multi-agent compliance systems for NIST 800-171, autonomous browser agents for airline seat allocation using VLM-based UI detection, end-to-end computer vision pipelines for sports analytics, causal ML pricing engines for banking, content recommendation engines for media publishers.

## Industries Served
- FMCG & Beverages: on-shelf availability prediction, stockout root cause classification, proactive alerting
- Manufacturing: hierarchical demand forecasting, production scheduling, ERP integration
- Fashion & Retail: sales forecasting, inventory optimization, financial reconciliation, production scheduling
- Banking & Finance: treasury forecasting, causal ML pricing, regulatory compliance
- Travel & Aviation: autonomous seat allocation, booking optimization
- Media & Publishing: personalized content recommendation, reader LTV optimization
- Cybersecurity / RegTech: multi-agent compliance automation, NIST 800-171 assessment
- Sports Technology: computer vision pipelines, player tracking, game analytics
- Retail Distribution: conversational analytics, demand forecasting, automated replenishment

## Team
- Luís Roque, Founder & Partner. Founder of the AI ventures TUTAI, Nixar, and Liftter. VP of Data & AI at Marley Spoon. Co-founder and CEO of HUUB (acquired by Maersk). Author of the "Deep Learning for Time Series Cookbook" (Packt). Published at AAAI and KDD, two of the top AI research conferences worldwide. PhD researcher at the AI Lab of the University of Porto.
- Rúben Pedrosa, COO. 8+ years in B2B sales and partnerships across e-commerce and fashion. Former roles at Platforme, ebankIT, and Fast White Cat. Runs go-to-market and enterprise client relationships across 11 industries.

## Ventures (ZAAI Ventures portfolio)
- TUTAI: AI education platform training professionals to build and ship AI (https://tutai.ai/en)
- Nixar: creator marketing platform for direct-to-consumer brands (https://nixar.ai)
- Liftter: business platform for personal trainers (https://liftter.com)
- HUUB: logistics platform for fashion brands, acquired by Maersk (exit)
- New ventures can apply at https://zaai.ai/ventures

## Deployment Evidence
- Median deployment time: 12 weeks
- 2,000+ concurrent time series managed
- 135,000+ transactions automated per year
- 42 store locations served in a single deployment
- 110+ regulatory controls assessed automatically

## Methodology
Five-phase delivery: (1) Requirements and architecture (weeks 1-2), (2) Data analysis and feature engineering (weeks 3-6), (3) Model development and benchmarking (weeks 7-12), (4) Testing, validation, and monitoring (weeks 13-16), (5) Knowledge transfer and stabilization (weeks 17+). De-risking approach: research phases validate architecture before full development commitment. Phased contracts with gated milestones.

## Contact
- Website: https://zaai.ai
- Email: hello@zaai.ai
- Book a demo: https://zaai.ai/contact

## Links
- Horizon: https://zaai.ai/products/horizon
- Pliney: https://zaai.ai/products/pliney
- Custom AI Solutions: https://zaai.ai/solutions
- Case Studies: https://zaai.ai/case-studies
- Blog: https://zaai.ai/blog
- Ventures: https://zaai.ai/ventures
- About: https://zaai.ai/about