29 days ago
Kyiv, UkraineSenior
Responsibilities
- Architect a unified GenAI gateway with automated routing, quota controls, and resilient fallback systems across model vendors.
- Improve system efficiency and latency through traffic management, response caching, token budgeting, and cost allocation controls.
- Build evaluation mechanisms for model-output precision, stability, and safety.
- Deliver distributed tracing, performance telemetry, cost analytics, monitoring, alerting, and observability across GenAI lifecycles.
- Standardize prompt engineering workflows with versioning, testing, governance, and production delivery integrations.
- Implement platform guardrails and safe production defaults in partnership with Security and Privacy teams.
- Participate in incident response, postmortems, monitoring, and a paid on-call rotation.
- Write production code, shape platform abstractions and operational standards, and collaborate with Applied Science, Data, Product, Security, and Engineering teams.
Requirements
- 5+ years of experience designing, building, and operating backend services or ML infrastructure handling production traffic.
- Strong Python skills and experience building developer-facing APIs, SDKs, or shared libraries.
- Practical experience with GenAI systems, including vector retrieval and external provider APIs.
- Hands-on experience using agentic AI and establishing quality gates and safety nets for AI-assisted engineering workflows.
- Knowledge of distributed systems, including high availability, concurrency, caching, and rate limiting.
- Experience with at least one major public cloud and its services and infrastructure, preferably AWS.
- Hands-on experience with distributed tracing, system telemetry, SLOs, and incident triage.
- Strong communication skills, at least C1 English proficiency, and willingness to join a paid on-call rotation.
- Experience operating AI gateways or scalable model-serving infrastructure is preferred.
- Familiarity with evaluation suites, prompt management, vector databases, or customized GenAI observability tooling is preferred.
- Experience with Metaflow, Databricks, or AWS SageMaker is preferred.
- Knowledge of hosting open-source foundation models, optimizing GPU compute, dynamic scaling, infrastructure cost controls, agentic workflows, multi-agent frameworks, or tool-use patterns is preferred.
Benefits
- Generous monthly allowance for lessons on Preply.com.
- Learning and Development budget, including time off for self-development.
- Competitive financial package with equity and leave allowance.
- Opportunity to support learners and tutors across more than 175 countries.
Categories
About Preply
Preply builds an online marketplace that connects learners with human tutors for live language lessons, supported by AI tools. Consumers book 1:1 sessions and ongoing courses, while organizations use Preply Business for corporate language training; revenue comes from commissions and subscriptions. Founded in 2012 and privately held, Preply operates globally with 100,000+ tutors teaching 90+ languages.