Remote AI Architect: Multimodal, High-Stakes Systems

Vor Systems, Inc.; San Francisco (CA); Remote; USD 140,000 - 190,000; Full time

Benefits offered by this job

Founding equity

Fully remote

In-person time as needed

Unlimited PTO with minimum of two days

Job summary

Vor Systems, Inc. is building AI-powered software to accelerate capital deployment and project development in renewable energy. You will join an early AI team to design and implement core reasoning systems and agentic workflows that plan, execute, and verify complex analyses, with emphasis on correctness and reliability.

This role is fully remote across the US, with optional in-person time as needed. We value ownership, strong Python/full-stack skills, and the ability to turn messy data into

Qualifications

• Production-grade LLM systems with agentic orchestration
• Python expert with full-stack experience
• Able to turn messy data into structured, usable insights
• Reliability-focused with observability and failure analysis

Responsibilities

• Architect the core AI reasoning system and agentic workflows that plan, execute, and verify complex regulatory, financial, and spatial analyses.
• Build multimodal intelligence pipelines on Modal to uncover critical information in messy sources like scanned leases, permits, and engineering schematics.
• Integrate rigorous validation and evaluation frameworks into high-stakes workflows where correctness and reliability are paramount.
• Own AI-powered features end to end, from models and backend workflows to the interfaces customers use to review results and take action.

Job description

Vor Systems, Inc. is building AI-powered software to accelerate capital deployment and project development in renewable energy. You will join an early AI team to design and implement core reasoning systems and agentic workflows that plan, execute, and verify complex analyses, with emphasis on correctness and reliability.

This role is fully remote across the US, with optional in-person time as needed. We value ownership, strong Python/full-stack skills, and the ability to turn messy data into