Engineering experience.
Personal responsibility.

InnerView AI is founded by Rob Royce, whose engineering experience includes NASA’s Jet Propulsion Laboratory and Matter.

That experience informs a practical approach: understand the constraints, ask what useful looks like, and stay close to the work of making it real.

NASA JPL and Matter are part of Rob’s individual professional experience.

Prior work / Engineering you can inspect

Engineering, put into practice.

Concrete examples from
Rob’s prior work.

Prior engineering / Matter Intelligence

AI in the workflow. AI in the product.

An AI-native development workflow.

At Matter, Rob built a development harness for the flight software team across Cursor, Claude, and Codex. It combined code-review agent skills with quality gates for generated code, making AI assistance part of a reviewed engineering workflow.

Merge-request throughput: 5×
Mean merge-request open-to-close time: 4 days → 5 hours

Workflow results reported in Rob’s résumé.

Natural language, connected to geospatial tools.

At Matter, Rob built an AI-native geospatial platform that made navigation, search, and data-layer controls available as agent tools. Users could interact with those capabilities through natural language. Rob demonstrated the platform to prospective utility and insurance customers.

Read Rob’s experience

Prior engineering / Open-source research

ROSA — Robot Operating System Agent

01Natural-language request
02Agent and defined tools
03ROS1 / ROS2 system

A simplified architecture

Robot operators need to inspect, diagnose, and operate complex systems. ROSA connects natural-language requests to tools for ROS1 and ROS2, giving an AI agent a defined way to interact with robot software.

Rob Royce is first author of the ROSA paper, developed with collaborators at NASA’s Jet Propulsion Laboratory. The open-source code makes the engineering inspectable.

The paper reports demonstrations with NeBula-Spot in JPL’s Mars Yard, EELS in a laboratory, and Carter in simulation. These are small-scale proofs of concept; the paper calls for further study.

The relevant engineering challenge: connect a model to useful tools, make its actions explicit, and test it against the system it needs to work with.

Inspect the code Read the paper

These examples come from Rob’s prior work at NASA’s Jet Propulsion Laboratory and Matter, with collaborators. They are not InnerView client engagements or institutional endorsements.

Let’s start with your priority

Which AI priority
needs to move forward?

Tell us what needs to change, what you have tried, and where you are stuck. The first conversation is to understand the problem and decide whether a focused piece of work makes sense.

hello@innerview-ai.com

Engineering experience.Personal responsibility.