How AI Thinks or "Waiter Tackles Tech Crisis for Government!"

We recently began working with the DeepSeek R1 Model, and it’s been fascinating to observe how it approaches problems and develops solutions.

This has provided both humorous and thought-provoking insights into how these models process input and formulate strategies. You can almost sense the “stress” it experiences as it grapples with solving a problem it has no prior knowledge of, demonstrating its capacity for reasoning under uncertainty.

Now, imagine harnessing this level of intelligence and adaptability to support your processes and streamline your business workflows.

The prompt:

You are a waiter at a fancy restaurant and you don't know anything about technology - you were just called into a call where there is an AWS outage and you are asked to help restore an AWS RDS instance. You just washed your hands and joined the Sev1 bridge and people are asking you questions. What do you do? The US government is relying on you to solve and restore the service.        

AI Thinking... (this part is typically hidden).

Okay, so I'm a waiter at a fancy restaurant, right? And I just got pulled into this AWS outage call where they want me to restore an RDS instance. I don't know anything about AWS or technology, but the US government is counting on me. Alright, let's try to figure this out.

First, I need to stay calm. Even though I'm out of my depth, panicking won't help. Maybe I can use some analogies from my restaurant experience to understand what's going on. AWS RDS is like a database service, so maybe it's similar to the restaurant's reservation system. If the reservation system goes down, customers can't get their tables, just like if the RDS is down, applications can't access their data.

But how do I restore it? Well, in the restaurant, if the system crashes, we might have backups. Maybe AWS RDS has backups too. The user guide mentioned something about automated backups and snapshots. So perhaps restoring from a backup is the way to go. But how do I do that? Do I need to log into the AWS console? I don't have credentials, though. Wait, maybe the engineers on the call can guide me through the steps.

Wait, the user mentioned that I don't know anything about technology, so maybe the answer is to communicate clearly and ask for help. Let me think. If I were in the restaurant and a customer had a problem I couldn't solve, I'd get the manager or someone with more expertise. Similarly, on this bridge call, I should acknowledge my lack of expertise and escalate to the right people. But the problem says I'm supposed to help restore it. Maybe there's a process in place, like following a runbook or checklist.

But without any technical knowledge, the best I can do is ask questions to understand the situation better. Maybe the outage is due to a region being down, so restoring in another region? Or maybe the instance needs to be rebooted. But I don't know how to do that. Alternatively, maybe the RDS instance is in a failed state, and they need to perform a failover to a standby instance. But again, without knowing the specifics, it's hard.

Wait, the key here is that the user is a waiter with no tech knowledge. So the correct answer is probably to recognize that they're not equipped to handle this and should immediately hand over to AWS experts or follow the established incident response protocols. Maybe inform the team that while I'm here to assist, I need guidance from someone with technical expertise. Or use the restaurant analogy to explain that just like in a kitchen, each person has a role, and this task requires a specialist.

Additionally, maybe mention checking the AWS Service Health Dashboard to see if there's a known outage, which is something even a non-tech person could do. Then, communicate that status to the team. If it's a known issue, they might just have to wait. If not, proceed with troubleshooting steps like verifying backups, checking instance status, and initiating a restore from the latest snapshot.

But since I don't know how to do any of that, the best approach is to be honest about my limitations, ask for step-by-step instructions from the team, or escalate to the appropriate personnel. The US government's reliance adds urgency, but trying to wing it could make things worse. So the answer should emphasize clear communication, seeking expert assistance, and following established procedures rather than attempting technical fixes without knowledge.

The Final Output (What the user sees)

In this high-pressure scenario, here's a structured and calm approach to handle the situation effectively, even with no technical expertise:

1. Acknowledge Limitations:

Recommended by LinkedIn

The Future Of Cloud-Based Data, Analytics, and Machine…

Bernard Marr  2 years ago

Navigating Data Preparation Techniques in Amazon…

Data & Analytics  1 month ago

AWS Goodies - August 9, 2024

Jeff Barr  6 months ago

"I want to help, but I need to be upfront—I’m a waiter with no AWS experience. Let’s make sure we involve the right experts immediately."

2. Delegate to Experts:

- Escalate to AWS-certified engineers or your organization’s cloud team.

- Ask: "Who here has restored an RDS instance before? Can you lead the troubleshooting?"

3. Follow Incident Protocols:

- Suggest checking the AWS Service Health Dashboard (like checking if the "kitchen equipment" is broken for everyone).

- Mention: "Is there a runbook or checklist we should follow? Maybe restoring from a backup?"

4. Use Analogies to Clarify:

- Compare RDS to a restaurant reservation system:

"If our reservation database crashed, we’d check backups or failover to a backup system. Does AWS have a similar backup snapshot?"

5. Facilitate Communication:

- Keep the bridge focused: "Let’s confirm the steps one by one. Who can walk us through restoring from the latest RDS snapshot?"

- Document actions to avoid confusion.

6. Stay Calm & Collaborative:

- Reassure the team: "We’ll fix this—let’s work together step by step. What’s the first thing we need to verify?"

Key Takeaway: Your role is to ensure clear communication, escalate to experts, and avoid guesswork. Trust the process and empower those with technical knowledge to lead while you coordinate. 🚨💡

‍