Founding AI Systems Engineer — Agent Harness & Production

Goaly AI; Palo Alto (CA); Hybrid; USD 180,000 - 240,000; Full time

Benefits offered by this job

Hybrid in Palo Alto

Visa sponsorship

Meals and snacks

Job summary

Goaly AI in Palo Alto is seeking a seasoned software engineer to build the harness between frontier models and real users. You will own the agent loop, tool interfaces, sandbox environments, and the end-to-end rollout of thousands of agent instances.

You will ship in production, learn from runs, and evolve environments, evals, and models. This founding role offers ownership from day one with minimal playbooks and high-impact outcomes.

Qualifications

• 5+ years building production software at scale.
• Experience shipping an agentic system with real users.
• Experience with failure modes: context overflow, looping, prompt injection, runaway cost.

Responsibilities

• Own the harness: the agent loop, tool interface, system prompts, permissions, stop conditions, retries, budget caps, and output checks.
• Build sandboxed execution environments (containers / microVMs) for thousands of rollouts in parallel.
• Design context engineering: window content, memory, retrieval, and sub-agent hand-offs.
• Make every agent run observable and replayable with traceability of calls and state changes.
• Build the eval stack: offline suites, production traces, and automated checks.
• Ship with customers: move prototypes to stable production and create reusable blocks.
• Close the loop with research: convert traces into tasks and reward signals.
• Make model calls: decide where each model runs and manage cost/latency/quality.

Skills

5+ years production software

Python

TypeScript

Go

Rust

Docker

Linux

Cloud infrastructure

Agent-based systems

Tools

Containers / microVMs

LLM integration

Job description

Goaly AI in Palo Alto is seeking a seasoned software engineer to build the harness between frontier models and real users. You will own the agent loop, tool interfaces, sandbox environments, and the end-to-end rollout of thousands of agent instances.

You will ship in production, learn from runs, and evolve environments, evals, and models. This founding role offers ownership from day one with minimal playbooks and high-impact outcomes.