AI lab researching human-agentic interaction and building tools for people and agents to work together
AI agent platform to automate and monetize customer service
Stage
Not given
Not given
Founded
Not given
Not given
Based in
DE
LT
Pricing model
Freemium
Not given
Pricing
Pirol AI: Buy budget starting from €20 with no subscription or expiration; priced per model from €0.62/1M output tokens for Pirol AI's blended rate. GLM-5.3-Flash: €0.19 input, €0.62 output per 1M tokens. DeepSeek V4.1 Flash: €0.40 input, €1.59 output per 1M tokens. GLM-5.3: €1.47 input, €5.45 outpu
Not given
Key features
Desktop app for local AI agents
Automatic CRM, ERP, HR software creation
Available for macOS and Windows
EU-hosted models with OpenAI-compatible API
Company as Code markdown standard
Model selection and routing
Multiple model options including vision
No subscription required
Budget does not expire
Multi-step case resolution including refunds, subscription changes, order disputes
Native integrations with helpdesk, CRM, order management, billing, and payment tools
Automated escalation to humans for unclear or sensitive cases
Revenue optimization through churn prevention and upselling
Context-aware natural conversation across channels
Support across full customer journey from pre-purchase to retention
Voice, chat, email, and social channels in 90+ languages
Complete decision transparency and traceability logging
Agent Graph visual system with real-time coordination
Forward-Deployed delivery with embedded engineers
What makes it different
All processing stays on your machine with no cloud uploads required
Agents can be trusted with sensitive work because they are predictable and observable
European hosting and data sovereignty
Open-source Company as Code standard for describing organizations
Model router picks best model per task
Resolves up to 87% of support cases and cuts costs up to 60%
Live in weeks, not quarters with no engineering team required
Pre-trained agents arrive with industry knowledge
Full visibility with no black box decisions
Agents reason within guardrails instead of following rigid scripts