Bayu Permana

Self proclaimed developer

[Google Gemma @googlegemma]

Meet DiffusionGemma!

An experimental open model that explores a fast approach to text generation, released under an Apache 2.0 license.

Moving beyond sequential, token-by-token processes to generate entire blocks of text simultaneously. Here’s what’s new with DiffusionGemma: 👇

[Aidan @aidan0x13]

HAHAHAHAHAHHAHAH

[Teknium 🪽 @Teknium]

Hermes Agent just surpassed VSCode in stars 🤩

Thanks everyone!

If you havent checked out the repo yet, check it out here:

https://t.co/EiLSJJCkGb

[AshutoshShrivastava @ai_for_success]

🚨 Google introduced a new Agentic RAG framework for Gemini Enterprise Agent Platform that makes AI systems much better at retrieving and reasoning over enterprise data.

Most traditional RAG systems work in a very simple way: they search once, grab a few documents, and generate an answer.

That approach breaks down when information is scattered across multiple systems or databases.

Google built a multi agent RAG system that can:
> Plan multi step searches
> Rewrite queries dynamically
> Search across multiple data sources
> Detect when important information is missing
> Continue searching until it has enough context

Google calls this approach “Agentic RAG.”

The biggest innovation here is something called the “Sufficient Context Agent.”

Instead of blindly generating an answer from partial information, this agent checks whether the system actually has enough context to respond properly.

It evaluates:
> What information was retrieved
> What information is still missing
> Whether another search is needed

If the answer is incomplete, the system keeps searching instead of hallucinating or giving up too early.

The architecture includes multiple specialized agents working together:

1. Orchestrator
Handles the overall workflow and task delegation

2. Planner Agent
Breaks complex questions into smaller retrieval steps

3. Query Rewriter
Converts vague prompts into optimized search queries

4. Search Fanout Agent
Searches across multiple sources simultaneously

5. Sufficient Context Agent
Verifies completeness and requests additional retrieval if required

6. Synthesis Agent
Generates the final grounded response

Google says the system achieved:

> Up to 34% higher accuracy compared to standard RAG
> 90.1% accuracy in cross corpus retrieval tasks
> Nearly identical latency despite the added reasoning steps

The bigger takeaway is that enterprise AI systems are slowly moving from simple retrieval pipelines toward systems that can reason, verify, and iteratively search before answering.

This could significantly reduce hallucinations and make AI systems much more dependable for real enterprise workflows.

The feature is currently available in public preview inside Gemini Enterprise Agent Platform.

[spooli 🛸 🦆 @slavalapasta]

Bro, lock in.
Bro:

[Vaibhav Sisinty @VaibhavSisinty]

I just found out OpenAI gives you $50,000 in free API credits if you do one thing in settings. 🤯

It's called the Data Sharing Program. No free trial exists for the API. But this does.

Go to your OpenAI Dashboard → Data Controls → Sharing. Opt in.

You get $50,000 in credits for their latest models or 2.5 million tokens for the rest.

The catch: your data gets used by OpenAI for training and improvements.

So don't use this for client work or anything sensitive. Use it for learning, side projects, and experiments.

[Lummox @Lummox_eth]

HERMES AGENT CROSSED 140,000 GITHUB STARS IN 3 MONTHS AND JUST BECAME THE MOST USED AGENT IN THE WORLD.

Most AI agents forget everything between sessions.

Hermes writes its own skills from experience.

Next time it runs the skill, improves it, and gets faster.

Independent benchmarks show agents with 20+ self-created skills complete similar tasks 40% faster than fresh instances.

Qwen 3.6 where the 35B version outperforms last year's 120B models at one third the memory footprint.

DGX Spark with 128GB unified memory running everything locally at $0 per month after hardware.

The setup takes 30 minutes. LM Studio plus Qwen 3.6 27B for the model server.

One install script for Hermes. One config connecting them.

Set context window to 65,536 tokens or nothing works.

After one month of daily use your skills directory has 20 to 50 learned workflows.

Your Hermes is genuinely different from anyone else's.

[AVB @neural_avb]

Wake up babe THE Claude Code guy dropped a free masterclass on harness engineering

[Bayu Permana @4beyu]

@antigravity btw if any antigravity dev asking for the debug info :

[Bayu Permana @4beyu]

anyone getting this error on @antigravity 2.0 ??

[Hanif | AI For Productivity @hanifproduktif]

Bansos ChatGPT Pro ($100/month) GRATIS selama 6 bulan khusus untuk yang punya/maintain repo open source

Lumayan banget nih mumpung dollar lagi naik jadi harga langganan makin mahal 😭

[monokern @monokern]

https://t.co/RmlXgqYeJg

[Farza 🇵🇰🇺🇸 @FarzaTV]

Watch me control my computer with just my voice. This is the future of operating systems.

No hands.

GPT-Realtime 2.0 is very, very underrated.

Demo:

[Akshay 🚀 @akshay_pachaar]

Hermes agent masterclass.

In this video, I cover everything you need to understand and customize Hermes Agent. Self-evolving skills, three-tier memory, GEPA optimization, and going from 1 to 10 agents that work for you 24/7.

Enjoy!

Chapters:

00:00 - Intro
02:03 - How to get the most out of this video
02:32 - What we're building (and why it's wild)
07:11 - How the whole thing works under the hood
09:27 - The SOUL.md: your agent's personality file
11:15 - The 3-tier memory system that keeps it all together
14:16 - Skills: what your agent can actually do
16:49 - The self-evolving loop (agents that improve themselves)
19:58 - The curator: Hermes' built-in garbage collector
22:56 - GEPA optimization: making your agent sharper
25:08 - Installation and setup
27:38 - Connecting your agent to Telegram
30:36 - Configuring programmer with Claude Code
31:53 - Adding new skills (from a hub of ready-made skills)
34:59 - Going from 1 to 10 agent profiles
36:49 - Building a custom designer from scratch
40:42 - Anatomy of the .hermes folder (where everything lives)
45:05 - Skill taps: sharing skills via a GitHub repo
45:59 - Skill bundles: stacking skills for workflows
47:19 - Hermes Kanban (coming soon)
48:05 Outro

Cheers! :)

[Bayu Permana @4beyu]

@kimmonismus yes, it looks like the case

[Prof. Dr. Kaan Yılancıoğlu @drkaanyl]

Efsane bir egzersiz seti yapabilen için gerçek bir güç dayanıklılık testi. Spor geçmişiniz yoksa denemeyin derim.
Şu egzersizle gelişmeyecek kas, kardiyovasküler sistem olamaz…

[Bayu Permana @4beyu]

@itsamarl @citifes @iniamajuga ini bare minimum kocakk

[Hanif | AI For Productivity @hanifproduktif]

Lo bayar BPJS Kesehatan tiap bulan. Tapi lo cuma tau bisa klaim rawat inap. Ada manfaat lain yang masuk dalam iuran BPJS yang jarang orang tau:

1. Kacamata. Peserta BPJS dapat subsidi kacamata plus lensa setiap 2 tahun sekali. Rp150rb (kelas 3), Rp200rb (kelas 2), Rp300rb (kelas 1). Caranya: minta surat rujukan dari puskesmas tempat lo terdaftar, bawa ke optik yang bekerja sama dengan BPJS.

2. Scaling gigi gratis. Pembersihan karang gigi bisa diklaim di 1 kali per tahun. Di klinik swasta ini bisa kena Rp150-250rb. Banyak yang gak tau ini sudah termasuk dalam manfaat BPJS dasar.

3. Tambal dan cabut gigi juga termasuk. Di faskes tingkat pertama, gak perlu rujukan. Langsung datang ke puskesmas atau klinik rekanan BPJS tempat lo terdaftar.

4. Fisioterapi. Dengan surat rujukan dari dokter, lo bisa dapat sesi fisioterapi gratis. Biasanya 6-8 sesi per kondisi medis. Di klinik fisioterapi swasta ini Rp150-300rb per sesi.

5. ANC untuk ibu hamil. Pemeriksaan kehamilan gratis selama masa kehamilan di puskesmas. Termasuk cek tekanan darah, berat badan, dan konsultasi dengan bidan. Beberapa faskes juga cover USG.

6. Persalinan normal di faskes BPJS tidak dikenakan biaya. Untuk SC (caesar), bisa dicover kalau ada indikasi medis dari dokter. Bukan atas permintaan sendiri.

Cara klaim semua ini: mulai dari faskes tingkat pertama (puskesmas atau klinik) tempat lo terdaftar. Kalau lupa terdaftar di mana, cek di aplikasi Mobile JKN atau website BPJS.

[Tesla Owners Silicon Valley @teslaownersSV]

Engineering is magic