Customer Story
How a European e-health leader built encrypted, resilient messaging for sensitive patient data on Akka — 2,500 messages per second at peak, delivered by two engineers.
Doctolib built a distributed, encrypted messaging system that 1M+ healthcare providers rely on to track patient care, handle referrals, and exchange medical expertise. It sustains 2,500 patient data messages per second at peak, holds 99.9999% availability, and was delivered in three months by two engineers — with no external cache or message queue to operate.
Akka ROI Scorecard
from start to production, built by two engineers. Secure inter-server messaging shipped without stitching together an external cache or message queue.
engineers run the entire backend. Sharding, clustering, and replication are handled by the platform, removing the infrastructure Doctolib would otherwise staff and pay for.
healthcare providers served, with thousands of new users added every month and 2,500 patient messages per second sustained at peak traffic.
Business outcome: 99.9999% service availability for sensitive patient data.
Doctolib needed to modernize how healthcare providers track patient care, handle referrals, and exchange medical expertise. The messaging that carries this work moves sensitive patient data, so it had to be encrypted end to end, provably reliable, and compliant with European health-data regulation. Any interruption or data loss would put both patient care and regulatory standing at risk.
Meeting those requirements the conventional way meant assembling and operating an external cache and a message queue, then securing every hop between servers. That infrastructure carries a standing operational cost and a large compliance surface — precisely what a small engineering team cannot afford to run around the clock.
Doctolib built its messaging on Akka, using sharding, pub/sub messaging, clustering, and replication to move encrypted messages between servers without an external cache or message queue. Removing those dependencies cut both the operational load and the compliance surface, which is why two engineers can run the entire backend.
This is the Never Fail guarantee: Akka handles clustering, resilience, durable in-memory state, and traffic steering, so the application does not have to. Jasper Aarts, Senior Staff Engineer at Doctolib, put it directly: "What's great about Akka is that it just works. It's saved us a lot of hours in terms of setup, and we are able to run our entire backend system with just two engineers."
Scale. 1M+ healthcare providers rely on the messaging system, and thousands of new users join every month. At peak traffic it sustains 2,500 patient data messages per second.
Real-time healthcare. Health professionals can instantly and securely discuss patients' cases and protocols with colleagues and peers through rich media messaging or voice and video calling without interrupting time-sensitive workflows.
Security and compliance. This solution maintains the highest of global standards and complies with the GDPR, E-Privacy, the NHS information governance, DCB 0129, ICO, ISO 27001, and NEN 7510, 7512, and 7513.
Cost to operate. Two engineers run the entire backend. Because sharding, clustering, and replication are handled by the platform — and no external cache or message queue has to be maintained — Doctolib carries far less infrastructure and compliance overhead than a system of this scale usually demands.
Speed to production. The system went from start to production in three months with two engineers, delivering encrypted inter-server messaging without custom infrastructure to build first.
Business impact. Service availability reached 99.9999%, the level required to carry sensitive patient data for referrals, care tracking, and clinical exchange across a national provider base.
The same secure messaging fabric now enables agentic AI on the clinical surface. On the Akka Agentic AI Platform, an AI agent can triage inbound patient messages, schedule appointments, and route clinical requests to the right provider — reading and writing across the encrypted channels Doctolib already runs. Because governance, encryption, and durable state are properties of the platform, agent actions on patient data stay inside the same HIPAA- and GDPR-grade controls that carry the 2,500 messages per second today, with a full audit trail of every routing and scheduling decision.
The platform that scaled Doctolib's messaging to 1M+ providers is the same platform that builds and runs agentic AI today. Enterprises build on the Akka Agentic AI Platform directly, or have a system delivered and operated through Akka Specify. Both provide the production guarantees Doctolib has run its patient data on at 99.9999% availability.
