Senior AI Agent Harness Engineer: Secure Runtime

SuperApp; San Francisco (CA); Híbrido; USD 230.000 - 315.000; Jornada completa

Ventajas ofrecidas por este puesto de trabajo

Flexible PTO

Medical, dental, vision insurance

401(k) with matching

Parental leave & fertility benefits

Therapy sessions covered

Wellness stipend

Lunch on Us

Descripción de la vacante

SuperApp is hiring a visionary AI engineer to lead the design and development of the Agent Harness, bridging foundation models with production-grade execution. You will implement state machines, secure sandboxes, and tool-calling middleware to safely run code, search the web, manipulate data, and query databases in real time.

You will work on security guardrails, performance optimizations, and mentorship, shaping a scalable, multi-model collaboration platform used across teams and products.

Formación

• 6+ years of professional software engineering experience, designing and operating distributed systems.
• 2+ years of production experience building or modifying agent execution environments or tool-calling frameworks.
• Expert proficiency in Go and Python with concurrent programming and scalable API design.
• Strong experience with containerization/virtualization (Docker, Kubernetes) and sandboxing untrusted code.
• Deep familiarity with message brokers (Kafka/RabbitMQ), Redis, and database design.
• Bachelor’s or Master’s degree in CS/engineering or equivalent practical systems experience.

Responsabilidades

• Architect the next-generation Agent Harness execution runtime and state management.
• Design secure, isolated execution sandboxes (Docker, gVisor, WebAssembly, microVMs).
• Build high-throughput APIs and integration layers (MCP) to connect to services, DBs, and web tools.
• Implement robust guardrails against prompt injection and data leakage.
• Optimize context handling, token usage, and latency in agent loops.
• Create design documents and mentor engineers across teams.

Conocimientos

Go

Python

Microservices

gRPC

API design

Distributed systems

Docker

Kubernetes

Security

Runtime isolation

Educación

Bachelor’s or Master’s in Computer Science/Engineering

Herramientas

LangChain

AutoGen

Model Context Protocol (MCP)

Docker

WebAssembly

Descripción del empleo

SuperApp is hiring a visionary AI engineer to lead the design and development of the Agent Harness, bridging foundation models with production-grade execution. You will implement state machines, secure sandboxes, and tool-calling middleware to safely run code, search the web, manipulate data, and query databases in real time.

You will work on security guardrails, performance optimizations, and mentorship, shaping a scalable, multi-model collaboration platform used across teams and products.