Cyrek D (VIZZ)

open source advocate
tweets/xs are always #imho
follows/likes/reposts are not necessarily #endorsements
now on

[thumbpark @eomjipark]

Happy New Year! 🐍 Just arrived in Zurich for @EthereumZurich! On Friday, I'll be giving a talk on lessons from Web2 open-source success that can be applied to Web3.

Looking forward to seeing you all!

[BladeoftheSun @BladeoftheS]

That's £200 EXTRA profit from every single Household in the UK.

Why is this acceptable?

[Miss Jo @therealmissjo]

Hooters appears to have lost sight of what their customers wanted.

Body positivity and “healthy at any size” people don’t look good in short shorts and vest tops.

Now Hooters is filing for bankruptcy.

Who would have thought it?

[SpaceXAI @SpaceXAI]

Developers: Soon, we'll release Grok 3, both standard and reasoning, and DeepSearch via our API. See our blog for model comparisons

https://t.co/Td0mjE58WN

[SpaceXAI @SpaceXAI]

This is it: The world’s smartest AI, Grok 3, now available for free (until our servers melt).

Try Grok 3 now: https://t.co/Tj0afLoxEz

X Premium+ and SuperGrok users will have increased access to Grok 3, in addition to early access to advanced features like Voice Mode

[Poonam Soni @CodeByPoonam]

OpenAI just got DeepSeek'd again.

A tiny startup based out of London just dropped the world's most capable AI agent for the web.

Here's EVERYTHING you need to know:

[Aurimas Griciūnas @Aurimas_Gr]

AI Agents 101: 𝗔𝗜 𝗔𝗴𝗲𝗻𝘁 𝗠𝗲𝗺𝗼𝗿𝘆.

In general, the memory for an agent is something that we provide via context in the prompt passed to LLM that helps the agent to better plan and react given past interactions or data not immediately available.

It is useful to group the memory into four types:

𝟭. Episodic - This type of memory contains past interactions and actions performed by the agent. After an action is taken, the application controlling the agent would store the action in some kind of persistent storage so that it can be retrieved later if needed. A good example would be using a vector Database to store semantic meaning of the interactions.

𝟮. Semantic - Any external information that is available to the agent and any knowledge the agent should have about itself. You can think of this as a context similar to one used in RAG applications. It can be internal knowledge only available to the agent or a grounding context to isolate part of the internet scale data for more accurate answers.

𝟯. Procedural - This is systemic information like the structure of the System Prompt, available tools, guardrails etc. It will usually be stored in Git, Prompt and Tool Registries.

𝟰. Occasionally, the agent application would pull information from long-term memory and store it locally if it is needed for the task at hand.

𝟱. All of the information pulled together from the long-term or stored in local memory is called short-term or working memory. Compiling all of it into a prompt will produce the prompt to be passed to the LLM and it will provide further actions to be taken by the system.

We usually label 1. - 3. as Long-Term memory and 5. as Short-Term memory.

A visual explanation of potential implementation details 👇

And that is it! The rest is all about how you architect the topology of your Agentic Systems.

What do you think about memory in AI Agents?

#LLM #AI #MachineLearning

[Alex Veremeyenko @alex_verem]

I tested Grok 3 and DeepSeek v3 with same critical prompts.

The results will blow your mind.

Grok 3 Vs. DeepSeek v3

(Video demos are included)

[Christina Gabbitas @ChristiGabbitas]

If you're interested in learning more about my story, #VapingScars, commissioned by @NBBCouncil as part of their community safety initiatives funded by @WarwickshirePCC, here’s the link to listen, starting with Laura Williamson from @asthmalunguk: https://t.co/JBbpGHY85r
Huge thanks @gemmaoaten @RadioHumberside for making me so welcome today.🤗
#vaping #education #children #youngpeople

[Nick Heller @nick__heller]

@Bitwarden @brave @shivan_kaul What I really don’t like about brave is the crypto integration that can’t be turned off easily.

[Bitwarden @Bitwarden]

For the first time, @brave leads the browser category in the 5th Annual Data Privacy Week Survey! Catch @brave privacy engineering lead @shivan_kaul live now on this X space! https://t.co/z4YK8r78fU

[Ruben @runsthegame]

@JoJoFromJerz Canada and Mexico are implementing measures, seems like China will also make moves. Meanwhile, Elon and his DOGE team have been granted full access to the American payment system. The data mining jackpot. Americans shouldn't worry about that at all.

[Jo @JoJoFromJerz]

This bullshit trade war with Canada is a smokescreen for some nefarious fucking shit folks.

[Reform UK Exposed 🇬🇧 @reformexposed]

Sure Andrea, want to explain why you used to open solar farms (on real farms)?

[Alistair Canlin @alcanlin]

Reform are where they are in the poles because of the BBC
#bbclaurak

[ECP Safeguarding @ECPSafeguarding]

Join our FREE 20-min webinar on Feb 13th, 2025, to spot online scams & safeguard young people. For parents/carers & pros. Register to get tools, tips, and a recording if you can’t attend.

https://t.co/srvfyXXhtd

[ECP Safeguarding @ECPSafeguarding]

👋 Meet the Team: Monaza Dasu – Our amazing Business Administration Manager! Mona is passionate about creating safe spaces for children and vulnerable adults. Her HR and leadership background brings a unique mix of skills. 🌟 #ECPSafeguarding #TeamIntroduction #MakingADifference

[ECP Safeguarding @ECPSafeguarding]

Violence Against Girls and Women is a global crisis rooted in harmful norms and inequalities. Let’s break the silence and build a safer world.

Read our blog to learn how you can help.

https://t.co/ihQwHEbhxf

[ECP Safeguarding @ECPSafeguarding]

You can report a crime for someone else under certain conditions to ensure their safety. If the victim permits, professionals like doctors and teachers, or guardians can report on behalf of children. The aim is to protect vulnerable individuals and ensure crimes are addressed.