The problem with most AI agents is that they work alone. A marketing team might have one bot writing ad copy, another analyzing spend, and a third posting updates to Slack, but they don't talk to each other. Mindra, a startup founded in Istanbul last year, is betting that the real value lies in the orchestration layer that makes them collaborate [Crunchbase].
Its product, an agentic orchestrator, is designed to let specialized AI agents hand off tasks, share memory, and execute real actions across tools like Google Ads, Meta, and LinkedIn, all with an audit trail [Perplexity Sonar Pro Brief]. The founding team of three Koç University students closed a $1.2 million pre-seed round from TQ Ventures to build it, targeting enterprise marketing and finance teams with a freemium model that starts at $3,000 per month for paid plans [Mindra Blog] [egirişim, 2025].
A wedge in marketing automation
Mindra's initial focus is on marketing workflows. The orchestrator is designed to manage tasks like pausing underperforming ad campaigns, adjusting budgets based on real-time analytics, and notifying teams [Perplexity Sonar Pro Brief]. The $3,000 monthly entry point positions it as a serious tool for teams already spending significantly on ads and seeking to optimize labor.
The team behind the orchestrator
The technical founding trio brings a blend of applied AI engineering and product architecture experience. CEO Zeynep Yorulmaz was previously an AI engineer at Y Combinator-backed Mercura and held engineering roles at Kuika and Bosch [Crunchbase] [zeynep.io]. CTO İlker Yörü was a product architect at ControlWorks, and CPO Deniz Soylular was a co-founder at AINA [Crunchbase].
| Role | Name | Previous Role |
|---|---|---|
| CEO & Co-Founder | Zeynep Yorulmaz | AI Engineer, Mercura (YC W25) |
| CTO & Co-Founder | İlker Yörü | Product Architect, ControlWorks |
| CPO & Co-Founder | Deniz Soylular | Co-Founder, AINA |
The scale and skepticism test
For an orchestrator to prove its worth, it must handle complexity without introducing catastrophic failure. The system must maintain a consistent state across potentially long-running agent workflows. It needs robust error handling and rollback mechanisms for when an agent's API call fails mid-sequence. Finally, the audit trail must be immutable and detailed enough for enterprise compliance.
The sober assessment is that the greatest risk isn't technical novelty, but go-to-market execution against well-funded incumbents. The space for automating business workflows is crowded, and Mindra has not yet disclosed named customers or deployment case studies [Perplexity Sonar Pro Brief]. The next twelve months will be about proving that an orchestrator built by a small, technical team in Istanbul can reliably manage a Fortune 500 company's ad spend.