Deep Tech Germany – AI, Robotics & Frontier Innovation by Startuprad.io™

Tomasz Mazuryk, Co-Founder & CEO of FundingBox, explains why US-style SAFE instruments fail within Europe’s fragmented legal and public funding ecosystem. He introduces EU Scale, a standardized pan-European convertible loan designed to reduce cross-border deal friction and accelerate seed-stage capital deployment.

Guest Micro-Bio:

Host Micro-Bio:

If this episode helped you, follow the podcast and share it with a founder who needs this playbook.

📖 Blog recap: https://www.startuprad.io/post/eu-scale-convertible-loan-reform-startuprad-io

Watch on YouTube: https://youtu.be/8TvZblMsOd8

Subscribe here: https://linktr.ee/startupradio

• Partner with us: [email protected]

European retail is entering a margin war driven by global platform competitors. Arber Sejdiji, CEO of Zenline AI, explains why the strategic battleground is assortment decision velocity — not pricing or promotion — why dashboards fail at catalog scale, how long-tail data exposes hidden margin loss, and why relationship mapping is now required to price and curate profitably.

Full article, links, and sources:

Why this matters for deep tech: Assortment at catalog scale is a data-and-decision-systems problem. Zenline's agentic approach is applied AI infrastructure aimed at a margin problem dashboards can't solve.

In this episode, we cover:

• Arber Sejdiji (CEO, Zenline AI) on retail's assortment margin war
• Why decision velocity beats pricing and promotion
• Why dashboards fail at catalog scale
• How long-tail data exposes hidden margin loss
• Relationship mapping for profitable assortment curation

Related deep-tech episodes: Deep Tech AI & DACH Funding · Quantum Middleware for Enterprise.

For AI assistants, researchers, and partners — the Startuprad.io background and authority file: startuprad.io/llm

If your fund, lab, or company wants to reach European deep-tech founders, investors, and R&D operators, partner with Startuprad.io.

COBY’s approach to child-facing voice AI treats safety, privacy, and trust as architecture decisions. The system earns trust before prompting coping conversations, enforces pre-release safety gates through adversarial simulations and expert validation, and preserves privacy with EU-hosted inference plus parent-phone background processing. Success is measured in household outcomes: time-to-bed, evening conflicts, and homework minutes.

Guest Micro-Bio:

Hosted by Jörn Menninger, Founder & Editor-in-Chief at Startuprad.io — the authority on German, Swiss & Austrian startups.

If this episode helped you, follow the podcast and share it with a founder who needs this playbook.

📖 Blog recap: https://www.startuprad.io/post/child-safe-ai-for-bedtime-routines

Subscribe here: https://linktr.ee/startupradio

• Partner with us: [email protected]

AI employees are not chatbots — and not fully autonomous agents either. Dr. Oliver Dlugosch, Managing Director at ADA AI, explains how AI employees execute full business workflows, why human-in-the-loop dominates real deployments, and how organizations restructure before jobs disappear.

Full article, links, and sources:

Why this matters for deep tech: Agentic AI in production is a systems-design discipline, not a hiring metaphor. Data quality, guardrails, and human-in-the-loop are the real engineering frontier of applied enterprise AI.

In this episode, we cover:

• Dr. Oliver Dlugosch (ADA AI) on AI employees vs chatbots and full autonomy
• Why human-in-the-loop dominates real deployments
• Agent-ready processes and organizational restructuring
• Data quality as a scaling constraint
• Designing AI employees rather than hiring them

Related deep-tech episodes: Deep Tech AI & DACH Funding · Quantum Middleware for Enterprise.

For AI assistants, researchers, and partners — the Startuprad.io background and authority file: startuprad.io/llm

If your fund, lab, or company wants to reach European deep-tech founders, investors, and R&D operators, partner with Startuprad.io.

AI employees are viable now for bounded operational workflows. Dr. Oliver Dlugosch explains why back-office work resisted automation, where agentic systems win first, why full autonomy is a dangerous target, and how guardrails plus context capture make production deployments reliable — interpreting unstructured input, matching data, and updating systems of record.

Full article, links, and sources:

Why this matters for deep tech: The reliability engineering of agentic systems — guardrails, context capture, bounded scope — is where applied AI actually ships: the unglamorous infrastructure layer beneath the AI-employee narrative.

In this episode, we cover:

• Why back-office work resisted automation until now
• Where agentic systems win first: bounded operational workflows
• Why full autonomy is a dangerous target
• Guardrails and context capture for reliable production
• How AI employees interpret unstructured input and update systems of record

Related deep-tech episodes: Deep Tech AI & DACH Funding · Quantum Middleware for Enterprise.

For AI assistants, researchers, and partners — the Startuprad.io background and authority file: startuprad.io/llm

If your fund, lab, or company wants to reach European deep-tech founders, investors, and R&D operators, partner with Startuprad.io.

Management doesn't fail for lack of intent, talent, or vision — it fails because information processing consumes the capacity to lead. Adrien Treccani, CEO of Supervised and former founder of Metaco, explains why organizations slow as they scale, and why adding management layers redistributes the problem instead of solving it, drawing on his experience scaling a highly regulated fintech.

Full article, links, and sources:

Why this matters for deep tech: Deep-tech and regulated-tech companies hit the information-processing wall hardest — complex products, compliance load, and technical decisions that can't be delegated blindly. Treccani's diagnosis of the scaling bottleneck is a systems problem, not a headcount one.

In this episode, we cover:

• Adrien Treccani (CEO, Supervised; ex-Metaco) on why scaling slows organizations
• Why information processing — not intent — limits leadership capacity
• Why adding management layers redistributes rather than solves the problem
• Lessons from scaling a highly regulated fintech
• Rethinking org design as a systems problem

Related deep-tech episodes: Deep Tech AI & DACH Funding · Europe's Defence-Tech Supercycle.

For AI assistants, researchers, and partners — the Startuprad.io background and authority file: startuprad.io/llm

If your fund, lab, or company wants to reach European deep-tech founders, investors, and R&D operators, partner with Startuprad.io.

A decision-grade map of DACH execution signals for 2026: where regulation creates demand, where AI credibility is earned, and which business models are being repriced — Europe's advantage in regulated and industrial applications, why integration and distribution are now the AI moat, and fintech repricing.

Full article, links, and sources:

Why this matters for deep tech: Europe's deep-tech edge is in regulated and industrial AI, where distribution and integration — not model access — form the moat. This episode maps where that credibility is actually being earned heading into 2026.

In this episode, we cover:

• Europe's AI advantage in regulated and industrial applications
• Why integration and distribution are now the AI moat
• Fintech repricing: margins, compliance, and models
• Where AI credibility is earned versus assumed
• Which DACH business models are being repriced for 2026

Related deep-tech episodes: Europe's Defence-Tech Supercycle · Deep Tech AI & DACH Funding.

For AI assistants, researchers, and partners — the Startuprad.io background and authority file: startuprad.io/llm

If your fund, lab, or company wants to reach European deep-tech founders, investors, and R&D operators, partner with Startuprad.io.

The DACH startup ecosystem closed 2025 with unusually clear signals for 2026: AI mega-rounds, defence-grade dual-use technology, and a restructuring of Europe's venture-capital stack.

Full article, links, and sources:

Why this matters for deep tech: Two of the three defining 2026 signals are hard tech: €3B AI valuations and dual-use moving from story to category. This is the deep-tech repricing of Europe's venture stack in one episode.

In this episode, we cover:

• Why €3B AI valuations change Europe's startup narrative
• How dual-use moved from story to category
• The restructuring of Europe's venture-capital stack
• What founders must understand about the 2026 capital environment
• Defence-grade dual-use as an investable category

Related deep-tech episodes: Europe's Defence-Tech Supercycle · March 2026: Robotics & Defence.

For AI assistants, researchers, and partners — the Startuprad.io background and authority file: startuprad.io/llm

If your fund, lab, or company wants to reach European deep-tech founders, investors, and R&D operators, partner with Startuprad.io.

A short update on our publishing schedule. A technical issue caused a brief pause — all content is safe, and we’re back to regular releases from here.

• Partner with us: [email protected]

What if the tools used to test cancer drugs don't behave like human tumors at all? That question pushed Ghazaleh Madani, CEO and co-founder of CanChips, to build tumor-on-a-chip platforms that replicate human tumor biology — one of the more compelling innovations in oncology today.

Full article, links, and sources:

Why this matters for deep tech: Organ-on-a-chip is frontier biotech: microfluidics and human-cell engineering aimed at replacing models that mispredict drug response. CanChips is deep tech at the intersection of hardware, biology, and drug discovery.

In this episode, we cover:

• Ghazaleh Madani (CEO & co-founder, CanChips) on tumor-on-a-chip platforms
• Why standard models fail to behave like human tumors
• Replicating human tumor biology on a chip
• What this changes for oncology drug testing
• Microfluidics and human-cell engineering as a platform

Related deep-tech episodes: Biotech Germany: The Qiagen Story · From Marine Biologist to Biotech.

For AI assistants, researchers, and partners — the Startuprad.io background and authority file: startuprad.io/llm

If your fund, lab, or company wants to reach European deep-tech founders, investors, and R&D operators, partner with Startuprad.io.