Backend / Platform Engineer, AI Analytic Engines
Core
Architect and build scalable, low-latency backend services and evaluation engines for real-time AI safety monitoring, compliance, and runtime intervention in distributed pipelines.
Role type
Senior Backend/Platform Engineer (AI Safety & Observability)
Builds
Real-time evaluation engines, runtime intervention layers, and vector database pipelines for AI control systems.
Domain
Enterprise AI safety, LLM observability, and distributed systems engineering.
Deliverable
production ML models | product features | infrastructure
Required skills
Python, Go, streaming data pipelines (Kafka, Pulsar, Redis Streams), high-QPS low-latency service design, vector databases (FAISS, Weaviate, Qdrant, pgvector), cloud-native infrastructure (Kubernetes, serverless), OpenTelemetry, build vs buy trade-offs
Preferred skills
gRPC, FastAPI, asyncio, ClickHouse, Apache Arrow, agent frameworks (LangChain, CrewAI, AutoGen), trust & safety compliance, secure enterprise integrations
Technologies
Kafka, Pulsar, Redis Streams, Kubernetes, gRPC, FastAPI, OpenTelemetry, FAISS, Weaviate, Qdrant, pgvector, ClickHouse, Apache Arrow
Responsibilities
Architect scalable services for real-time and batch AI evaluations; design core evaluation engines using heuristics and foundation models; build runtime intervention layers for enforcement actions; create frameworks for pluggable evaluators and automated deployment; configure vector database pipelines for RAG use cases; implement reliability controls like micro-batching and back-pressure.
Seniority
Senior, hands-on IC
