I’ve always been fascinated by space not just the rockets and the visuals, but the systems that quietly keep humans alive beyond Earth The Moon, in particular, feels like the next great testing ground for human resilience: limited resources, delayed communication, and zero margin for error.
That curiosity became the seed for LUNAOPS an autonomous mission operations assistant inspired by the challenges astronauts will face during long-duration lunar missions under programs like Artemis.
LUNAOPS simulates how an onboard AI could monitor habitat conditions, assess mission risk, track trends over time, and generate concise status reports suitable for transmission back to Earth, even when real-time human oversight isn’t possible.
🛠️ How I Built It
The project is designed as a lightweight, realistic mission analyzer, focusing on clarity and reliability rather than flashy interfaces.
At its core, LUNAOPS
Ingests structured mission telemetry (power, oxygen, temperature, task status)
Evaluates mission risk using deterministic rules
Tracks historical state to detect trends (improving, degrading, or stable systems)
Generates a human-readable mission report
Optionally augments the report with an AI-generated summary when connectivity is available
The system intentionally supports two modes:
CONNECTED mode uses an LLM to generate mission summaries
AUTONOMOUS mode falls back to a deterministic onboard summary if AI is unavailable
This mirrors real space constraints, where communication delays or outages are expected.
All outputs are saved as Earth-transmission-ready reports, emphasizing low-bandwidth, high-signal communication.
🧠 What I Learned
Building LUNAOPS taught me lessons beyond just writing code:
How to design systems that fail gracefully instead of breaking
Why state history matters for decision-making, not just current values
How small implementation details (encoding, file handling, version control) can become critical
How to think like a mission engineer: What happens when something goes wrong?
⚠️ Challenges Faced
Some of the biggest challenges included:
Designing trend detection that distinguishes between no data, stable systems, and actual degradation
Handling Unicode and encoding issues across different environments
Managing an evolving project structure without breaking existing logic
Resolving Git conflicts under time pressure while keeping the project clean and shippable
Each challenge forced me to slow down, reason carefully, and build something more robust.
🌌 Why It Matters
LUNAOPS isn’t just a script it’s a conceptual prototype of how autonomous decision-support systems can assist humans beyond Earth.
As we move toward a sustained lunar presence, tools like this won’t be optional — they’ll be essential.
This project represents my first step toward building those systems. LINK TO GITHUB REPO https://github.com/wafae1116/lunaops-artemis/tree/main
Log in or sign up for Devpost to join the conversation.