The king is dead, long live the king!

Context Engineering is a new Vibe Coding. As we know, LLMs become pretty unreliable as Context grows. Industry came up with different ways to address it like Context Compression, Context Trimming ect. Folks are seeing much better results when Context data is engineered properly.

https://github.com/humanlayer/12-factor-agents

Hi folks. I am hiring Sr Staff Software Engineer. I would appreciate the repost:

https://www.linkedin.com/posts/bkodirov_hiring-activity-7351791921607946240-rbEF?utm_source=share&utm_medium=member_desktop&rcm=ACoAAAzP3TIB4pvnEUAvi_ABR6D5qCr_LxAIk0s

There is a really cool book about Agent Development got dropped today. Check it out: https://docs.google.com/document/d/1rsaK53T3Lg5KoGwvf8ukOUvbELRtH-V0LnOIFDxBryE/edit?tab=t.0

Come over if you want to hang out :)

31 июль, соат 19:00 да IMPACT.T да АҚШда логистика соҳаси учун AI агентлар яратаётган Numeo стартапи асосчилари — Beka Kodirov ва Otabek Yusupov иштирокида "AI стартаплар қандай қурилади?" мавзусида учрашув бўлиб ўтади.

Қуйидаги мавзулар муҳокама қилинади:
- Силикон водийсида AI стартапларини қуришда учрайдиган муаммолар
- Агент технологиялари қандай ишлайди?
- Voice AI имкониятлари
- Ўзбек дастурчилари ва фаундерлари учун АҚШ бозоридаги имкониятлар

Спикерлар:
Beka Kodirov (CTO) — Warner Bros. Discovery, Hulu ва fuboTV каби компанияларда 10+ йиллик тажрибага эга муҳандислик раҳбари булган. У Max ва Discovery+ учун Video SDK ишлаб чиқишни бошқарган. Ҳозирда Numeoда AI архитектураси ва маҳсулот стратегиясини бошқармоқда.
Otabek Yusupov (COO) — молия ва операцион фаолият соҳасида 14+ йиллик тажрибага эга. У 8 йил Deloitte ва 6 йил Amazon компаниясида ишлаган. Amazon’да глобал логистика жамоалари учун молиявий стратегия ва режалаштиришни бошқарган.

Жойлар сони чекланган — чипта харид килишга улгуринг!

@impactthub

Our current AI workflows keep facing new challenges. Manually updating workflows doesn't scale well.
I am researching around self learning agents lately.
I run into some interesting research papers that popped up lately.
I am reading the fllowing one currently.

https://arxiv.org/html/2508.00271v2

To: Media Nerds

FFMPEG 8.0 has landed 🚀!
https://www.linkedin.com/pulse/ffmpeg-80-huffman-unleashed-generational-leap-forward-lee-erickson-tmtme/?trackingId=WN8hQ20TxjW1xLLRMx2R7Q%3D%3D

- It brought good amount of Vulkan based Codec (Encoding and Decoding) support that brings amzing speed bump to your media processing workflow.
- There are new CUDA filters (a.k.a. transformers)

Did you know that Toyota has a footprint in modern GenAI frontier?
Toyota Research Institute is one of the folks who is pushing the boundaries of Large Behavioral Models. Yeap, you have not missheard it. LBM - large models that are very similar to LLMs. They are trained on motions in 3D space if I put it simply.
You can fine tune those models and build your own robots that focus on Kitchen tasks, the ones that are specialized in yard tasks ect.
That is one of the reasons why they say that 2026 is Robotics era kick off .

https://www.tri.global/our-work/large-behavior-models

How I accedentally discovered Super Power of Comet browser.

I wrote an ADR(Architecture Decision Record) with Windsurf Claude Sonnet to change the feature we already have. I got a bit lazy and wanted somebody to read the ADR outloud.
I opened numeo Github Repo on my belowed browser Comet to check if it could read the MD file outload for me. Comet suprized me immediately by identifying the PR page, and failing CI steps. That triggered me to think... "Could I ask Comet browser to review our PRs??

So I tried.... It started going through all the PRs one... by... one....

Topic: AI Agent reliability #agentreliability

MAKER, developed by Cognizant AI Lab with UT Austin, is an agentic system that completed a task with over one million steps and zero errors, addressing the reliability limits of current LLMs for long, chained reasoning. Unlike the usual focus on increasing model intelligence, MAKER uses extreme task decomposition: each tiny subtask is given to a dedicated microagent, and agent outputs are error-corrected by local multi-agent voting and red-flag filtering. This modular framework makes reliability scale logarithmically with the task length (Θ(s log s)), instead of suffering exponential error growth, enabling small and efficient models to outperform larger ones on extended tasks. The success of MAKER suggests that high reliability in large-scale AI can come from orchestration and structure, not just bigger models

Link: https://arxiv.org/abs/2511.09030v1

One of the most practical book to bulid RAG I have encountered lately. It shows challenges around RAG and how to address them step by step. One by one.

https://www.youtube.com/watch?v=-Tgc_9uYJLI
Thisis a big leap forward in On-Device / Edge infrence. This model can work directly on your devices or browser.

Our Agents @Numeo have to have lots of skills/instructions. This comes with its challanges. You probably have heard about "Curs of Instructions" paper that explains why having too much instructions is bad.

At Numeo we have been doubling on building our Agents in a way so they can "Pick Up" skills/instructions progressively. We will defenitly experiment with Function Gemma.
I am expecting our tool calls will be much more precise, much faster and much cheapper with this model.

We are living in a world where
- Resumes are generated by GenAI,
- Applications are submitted by AI agents
- Replies are written by AI agents :)

STEP-3.5-FLASH: Breaking the Speed vs. Intelligence Trade-off

StepFun released a model with 196 billion parameters, but only 11 billion activate per token. This is a sparse Mixture-of-Experts model. The model is performed pretty well on SWE-Bench. People are claiming that to be GPT-4 level quality which is pretty good based on my experience.

Its throughput is 300 tokens per second on modern hardware. Compare that to DeepSeek's 33 tok/s. You're looking at roughly 9x faster inference on the same task, with higher accuracy on coding problems.

Context window: 256K which is pretty cool. Handled through a 3:1 sliding window attention mechanism that avoids the quadratic complexity that kills most long-context models.

Why This Matters for Agents

The model uses multi-token prediction—predicting three tokens per forward pass instead of one. Sounds simple. It's not. This breaks the serialization bottleneck in autoregressive generation and fundamentally changes the latency profile for agentic workflows.
For autonomous agents, this matters. You need models that reason deeply but respond quickly. Step-3.5-Flash handles both without the usual trade-off.

Local Deployment

Runs on a Mac Studio M4 Max with 128GB. You get GPT-4 class reasoning entirely offline. That's a shift for privacy-sensitive work, edge deployment, or cost-critical inference.
INT4 quantization is available. Cuts memory further without catastrophic loss in capability.

Real Benchmarks Beyond Coding
🔗 GitHub: https://github.com/stepfun-ai/Step-3.5-Flash
🔗 Hugging Face: https://huggingface.co/stepfun-ai/Step-3.5-Flash
🔗 NVIDIA NIM: https://build.nvidia.com/stepfun-ai/step-3.5-flash
🔗 OpenRouter API: https://openrouter.ai/stepfun/step-3.5-flash
🔗 Original Blog: https://static.stepfun.com/blog/step-3.5-flash/

We are hiring from Uzbekistan

My laziness hit a new rock bottom :).

Now Claude Code is reading its output out loud:).
So I don't have to read small text out of my screen all the time :).

Local TTS
https://github.com/ktaletsk/claude-code-tts#

So true. There is not much MOAT in software anymore.

I bet we are gonna see lots of Software companies start adding Hardware products to create a defensibility very soon.

AGI is here folks! Holly cow!

Claude Opus 4.7 did it!

https://www.youtube.com/watch?v=kYkIdXwW2AE
If you are aware, Yann LeCoun has left Meta and actively researching/developing and investing in other startups that are trying to move frontier in NON-LLM directions. Yann isinvesting / betting on World Models (Generic name).
He thinks LLMs are not the best proxies of the way Human intelligence works.

On this video, Yann is sharing his thoughts on JEPE models. Problems of Video generation approaches. Why Autoregressive models are not great for Video gen cases.

Enjoy folks.