hjaveed

I’ve been seeing a lot of noise about Clawdbot, so I spent time actually running it and writing down what I learned.

2026 could be the year of personalized agent harnesses in a box, where developers use agent SDKs to build their own systems, specialized and verticalized for a specific use case.

Under the hype, the interesting part is simple: this is AI that lives with you, not something you open and close. It runs, remembers, and does work in the background. That shift matters more than the product itself. And yes currently it's a security nightmare, but we will evolve soon using more secure sandboxes to run the AI.

I wrote this as a thinking-out-loud piece, not a review.

Sharing here if you want to read: https://www.hadijaveed.me/2026/01/28/clawdbot-beyond-the-hyp...

the cost of creating a great product and amount of time it takes to get infront of the customers has still not reduced

good for the founding team! Kafka is an enterprise bloat. most of the queueing solutions could be built with something much simpler

this is super cool.

Inspired by the ideas from this Cognitive Architecture paper and an insightful Langchain Blog by Harrison Chase.

For production use-case, we're exploring how to create closed-loop, safe agents in healthcare — systems that can reason and execute on patient needs in a secure and reliable way. The key is understanding how these agentic systems should think, the flow of execution in response to patient intent, and ensuring safety through structured, observable loops.

We're inspired by the ideas from this Cognitive Architecture paper and an insightful Langchain Blog by Harrison Chase.

At RevelAI Health, we're exploring how to create closed-loop, safe agents in healthcare — systems that can reason and execute on patient needs in a secure and reliable way. The key is understanding how these agentic systems should think, the flow of execution in response to patient intent, and ensuring safety through structured, observable loops.

I have suffered from this myself. The only way forward to your point is to take action and not worry about the outcome or how others perceive it. I write now to share, document and also to make myself accountable

I have suffered from this myself. The only way to move forward is to take action and not worry about the end result or outcome. I write to share my ideas now, document and also to make myself accountable

No matter how innovative your healthcare product is, you won't get deployed without convincing the security team. Here's what I learned from successfully implementing HIPAA compliance at large health systems - focusing on practical cloud architecture decisions and navigating dated security questionnaires.

can you share any resource that mentions about teaching the model to do COT.. their release blog does not document much

For patch management particularly with EC2s, we use AWS Systems Manager Patch Manager.... fairly straightforward to setup once you configure a base image

obviously, it's not cloud-native... but if you are using AWS EC2 it works

After listening to @levelsio on Lex Friedman’s podcast, I became obsessed to simplify my deployments:

Do startups really need complex cloud architecture?

Inspired, I wrote a blog exploring simpler approaches and created a docker-compose template for deployment

Curious to know your thoughts on how you manage your infrastructure. How do you simplify it? How do you balance?