Priyatham Kattakinda

Co-founder & CTO @ReliableAI | ECE PhD @UofMaryland

[Priyatham Kattakinda @priyathamkat]

Excited to share that I've stepped into the role of CTO at @ReliableAI!

Grateful to my team and everyone who's helped along the way. Building reliable AI agents is what we're here for, and we're just getting started.

Special thanks to @FeiziSoheil and @py_laurent for their invaluable guidance and support.

[Priyatham Kattakinda @priyathamkat]

@sarahwooders Do you always leave it on/awake?

[Priyatham Kattakinda @priyathamkat]

@github actions down again. SMH.

[Priyatham Kattakinda @priyathamkat]

@DominickNoval All the experiments live in a private repo and every change after an experiment run gets committed. I have optimized the design for simplicity. The plan is to add sensible changes when the need arises. It is by no means a production ready design.

[Priyatham Kattakinda @priyathamkat]

Personal agents are great at chatting, but I wanted mine to do long-running work on my own code and data. So I built an MCP server that lets Duet launch Claude Code on Daytona sandboxes and show results accessible from my phone. Details in the article: https://t.co/Bd1JzZ4UGK

[Priyatham Kattakinda @priyathamkat]

@HashgraphOnline Thanks; will take a look.

[Priyatham Kattakinda @priyathamkat]

@HashgraphOnline Yes to both. Since the sandboxes are throwaway and don't have any secrets, I am comfortable letting claude work without supervision. Models today are good enough that, for my use cases, they need minimal steering.

[Priyatham Kattakinda @priyathamkat]

@drunk_vibecoder Lockfiles are committed but yes, big picture, I need a way to store results and other metadata when the scale of experiments grows. I haven't hit that point yet.

[Priyatham Kattakinda @priyathamkat]

@adammocklerr @grok how many of these were re-elected last week?

[Priyatham Kattakinda @priyathamkat]

@reasonmachines I built my own custom personal agent, how can I use @reasonmachines to connect you to it @bot ?

[Priyatham Kattakinda @priyathamkat]

@8teAPi How do you feel about personal agents having access to a lot of your private data?

[Priyatham Kattakinda @priyathamkat]

@FeiziSoheil Thank you so much @FeiziSoheil for the kind words! The feeling is mutual. I am grateful for having an amazing advisor and mentor!

[Priyatham Kattakinda @priyathamkat]

@thdxr How's Origin from Cursor?

[Priyatham Kattakinda @priyathamkat]

Why doesn't Claude or Codex support A2A natively?

[Priyatham Kattakinda @priyathamkat]

Update:

[Priyatham Kattakinda @priyathamkat]

@ChatGPT sends notifications on Mac even if there are no new turns in any of my chats. I think it’s coming from my dot, even if there are no new messages from it.

[Priyatham Kattakinda @priyathamkat]

@ChatGPT sends notifications on Mac even if there are no new turns in any of my chats. I think it’s coming from my dot, even if there are no new messages from it.

[Priyatham Kattakinda @priyathamkat]

@paahrsa Like a dæmon!

[Priyatham Kattakinda @priyathamkat]

Opus 5.5 is really really good. I say this as I use it to clean up obvious mess from a different model 👀.

[Priyatham Kattakinda @priyathamkat]

I have used many open-source agents as personal assistants including a couple that I built myself. I always found it too much trouble to adapt them to my needs. Something would always break and I would end up spending more time fixing things than having them do something useful for me.
Lately, I have been using @ChatGPT dot and I really like the UX. The only problem is that it doesn’t have all the features I need and I can’t control what features @OpenAI ships or how fast they do it.
So I went back to the drawing board and started building one from scratch taking inspiration from my dot. It has the ability to modify any and all of its code, learn new skills, etc. (not saying these are novel ideas, I know @openclaw, @NousResearch have something similar). So far, it hasn’t gone down by breaking itself despite modifying itself several times. There was only one instance where it made an edit that broke it but it was able to automatically rollback, fix that issue and redeploy. That really surprised me, pleasantly, of course! I think the models have gotten really good over the last few months. They will continue to get better at one-shotting tasks, recover from failures and keep pursuing long-term goals. Many of us have seen this coming. But building with the latest models, I feel like I am experiencing the inflection point firsthand.
There has never been a better time to be a builder!

[Priyatham Kattakinda @priyathamkat]

@artfreebrey @bot how does this work?