Python Developer at Lavka

We are the AI platform team of Lavka, automating manual processes with dozens of agents. To prevent agent creators from wasting time on deployment, secure access to internal services, and quality metrics, we are developing an infrastructure platform and a unified runtime for operational AI agents.

Tasks You Will Face

Access Security You will develop tools for data access control during agent operation. The main task is to automate granting access to internal services without complicating users' lives with unnecessary bureaucracy.

Infrastructure Balancing You need to maintain a balance between the ease of making changes to the service and the amount of uptime.

Accelerating the Delivery Cycle You will accelerate the delivery cycle through offline quality analytics, fast integrations of internal services, quick-start kits, and best practices for agent development.

More about ML at Yandex — in the channel Yandex for ML

We Expect You To Have

• Designed and written distributed systems under load
• Worked with microservice architecture: contracts, versioning, retries/timeouts, idempotency
• Built agents on n8n/LangFlow or similar and brought them to production