Agent Harness Engineer - Core Runtime
NinjaTech AI is a generative AI startup (B2C and B2B) with headquarters in Silicon Valley and offices in Sydney and Vancouver. Backed by Alexa Fund and Samsung Ventures, we are building the autonomous AI agents that let anyone get real work done with generative AI.
Our flagship product, SuperNinja, is an advanced agentic AI platform with full OS capabilities: website creation, end-to-end coding, advanced data analysis, and more. The three roles below build the agent harness itself - the runtime, integrations, and reliability layer that every SuperNinja agent runs on.
Position Overview
We are looking for a Core Runtime engineer to own the agent harness that turns our LLMs into reliable, autonomous software. The harness is the run loop, orchestrator, and monitor that keep an agent working safely across thousands of tool calls and hours of wall-clock time. You will work at the intersection of systems engineering and applied AI, building the substrate that every SuperNinja agent runs on.
Key Challenges & Opportunities
- Design and evolve the agent run loop: context management, turn compaction, and recovery from failed or partial tool calls
- Build the orchestrator that launches, supervises, and hands off long-running background agents without losing state
- Implement the monitoring layer that detects stuck, looping, or runaway agents and intervenes safely
- Own the provider abstraction across model backends (Claude, Codex, and others) so the harness stays model-agnostic
- Make the harness observable end to end with structured telemetry, so failures are debuggable in production
- Keep the loop fast and cheap at scale while it powers millions of daily user interactions
Experience & Education Requirements
- Master's or Bachelor's in Computer Science or equivalent practical experience
- 4+ years building production backend or systems software in Python
- Strong grasp of concurrency, process supervision, and fault-tolerant design
- Experienc...