AI Engineer: Self-Hosted Agent Orchestrator & Verifier

Pronoia; Torrance (CA); On-site; USD 130,000 - 210,000; Full time

Benefits offered by this job

Health insurance

Unlimited PTO

Relocation assistance

Professional development stipend

Access to cutting-edge labs

Job summary

Pronoia Energy, a pioneering quantum energy storage startup in Torrance, California, seeks an AI Engineer to build an internal, agent-driven system. You will orchestration plans, route tools, and implement the first verification gate to ensure factual grounding before the team acts on it.

You will deploy a self-hosted open-weights model on-site, create an evaluation harness, and enforce strict traceability to sources or deterministic computations, wearing multiple hats as a founding generalist.

Qualifications

• Experience shipping a real agent or tool-use system in production.
• Genuine experience self-hosting open-weights models on own GPUs, not just API calls.
• Treating hallucination and tool-misuse as engineering problems with mechanical fixes.
• Builder-generalist range across orchestration, model serving, and evaluation.
• Self-starter who thrives amid ambiguity and rapid iteration.

Responsibilities

• Build the agent orchestrator and the tool-routing logic at the heart of the system.
• Stand up and serve a self-hosted open-weights model on local hardware.
• Build the first verification gate and evaluation harness, grounding procedures to sources or deterministic computations.
• Enforce strict routing so claims trace to a retrieved source or a deterministic computation.
• Wear several hats as a founding generalist and later hand specialties to dedicated hires.

Skills

Agent orchestration

Tool routing

Self-hosting GPUs

Grounding & evaluation

Ambiguity handling

Tools

vLLM

TGI

Agent frameworks

Job description

Pronoia Energy, a pioneering quantum energy storage startup in Torrance, California, seeks an AI Engineer to build an internal, agent-driven system. You will orchestration plans, route tools, and implement the first verification gate to ensure factual grounding before the team acts on it.

You will deploy a self-hosted open-weights model on-site, create an evaluation harness, and enforce strict traceability to sources or deterministic computations, wearing multiple hats as a founding generalist.