Senior Machine Learning Engineer
RemoteMelbourne, Victoria, Australia
Job Summary
Architect fault-tolerant, multi-tenant agent runtimes featuring incremental execution, state persistence, and automated recovery for long-running workflows. Design high-throughput memory infrastructures supporting working, episodic, semantic, and procedural tiers with strict ACID transactions. Implement zero-trust execution validation and Row-Level Security to isolate tenant contexts within shared infrastructure pools. Drive adoption of the Model Context Protocol (MCP) by building secure servers and clients for dynamic tool discovery and Agent-to-Agent collaboration. Establish telemetry to trace agent actions, enforce rate limits, and optimize compute cost-per-token. Partner with AI Scientists to define technical direction for the platform's digital nervous system using Go, Java, or Python, leveraging event-sourcing and Kafka for distributed state management.
Required Qualifications
- Staff-level architect
- Extensive backend experience (Java, Go, or Python)
- Deeply understand the nuances of managing distributed state
- Deeply understand event-driven architectures (Kafka)
- Deeply understand non-deterministic execution in cloud environments
- Understand reliability in an AI system
- Understand externalizing state
- Understand safe self-correction
- Understand deterministic recovery
- Excited by open standards like MCP
- Understand extreme complexities of API design
- Understand capability negotiation
- Understand secure sandboxing for tool execution
- Advanced durable execution frameworks
- Event-sourcing orchestration
- High-throughput in-memory datastores
- Converged databases
- Vector stores
- Postgres
- Kafka
- MCP (Model Context Protocol)
- gRPC
- GraphQL
- Zero-trust execution validation
- Native Row-Level Security within the database layer
Desired Qualifications
- Infinite context windows
Hiring someone like this?
Get your role in front of qualified candidates on Sorce.