Nusret (@nsrt_py) header
Nusret (@nsrt_py) profile picture

Gamer | Reader | Hugging Face Builder

İstanbul, Türkiye
Joined December 2016
Photo by Nusret (@nsrt_py)Photo by Nusret (@nsrt_py)Photo by Nusret (@nsrt_py)Photo by Nusret (@nsrt_py)Photo by Nusret (@nsrt_py)Photo by Nusret (@nsrt_py)Photo by Nusret (@nsrt_py)Photo by Nusret (@nsrt_py)Photo by Nusret (@nsrt_py)Photo by Nusret (@nsrt_py)
Presenting Olala mini, a 7B total (1B active) parameter model trained on 13T tokens ! It uses a custom architecture mixing Mamba-3 with Grouped Differential Attention layers. My work focused on architectural improvements over the Transformer, which I detail below
14
43
6
381
21,726
Nusret retweeted
Atatürk'ün cenaze törenine dair pek çok kayıt ve fotoğraf gördüm. Ama beni en çok hüzünlendiren bu fotoğraf oldu. Anadolu'nun asil köylü kadınları, bebekleri kucaklarında tren yoluna çıkmış Atatürk'ün tabutunu Ankara'ya götüren treni bekliyor. Geçerken bir iki saniyeliğine de olsa görebilmek için... Vatanlarını düşmandan kurtarmış Halaskar'a son defa selam durabilmek için...
38
1,358
41
10,947
236,965
Atamızın son 300 günü. Yıllar önce ağlamaktan helak olduğum çok güzel bir belgesel.
16
295
30
2,526
103,191
Replying to @mhsmozr
Şu an bir Kaggle yarışmasıyla uğraşıyorum, yaptığım çalışmalar da daha çok onun için: kaggle.com/competitions/gemm… Yazı konusunda daha bir şey çıkaramam ama aşağıdaki kaynakları faydalı buldum: huggingface.co/spaces/FineEn… huggingface.co/blog/harness-… github.com/thinking-machines… Ayrıca ARC-AGI'den sonuçlar paylaşıldıkça onları okuyorum. Data toplama, speed up, fine-tuning vs. güzel insight'lar var. kaggle.com/competitions/arc-… kaggle.com/competitions/arc-…
1
3
16
509
Nusret retweeted
phd nefretini ve bu dramaların etrafında konumlanan insanları anlamakta çok zorluk çekiyorum. kimseyi hedef göstermeye çalışmıyorum ama acaba spesifik bir alandaki phd'li bir insanın o alanda sizden birkaç kat fazla "şey" bilmesi normal olabilir mi? ortak bir zeminde çok güzel tartışmalar dönebilir, çok da keyifli olur ama konuyla hiç alakası olmayan toksik twitter kişilerinin olaya taraftarlık yapması çok garip.
12
1
3
69
4,808
Nusret retweeted
Day 28/ tibo's game has landed i spent $5.99 on this domain, but i had a blast making it with grok bot and cursor! i hope you enjoy it and find all the easter eggs 🤍 if i did it right you should be able to play it inside of X too tibo.dog
1,506
886
838
14,066
2,259,455
Nusret retweeted
Your LLM eval is probably lying to you. Here are 4 ways it happens, and the cheap habits that catch each one. From my AI Connect Istanbul talk, with code you can run in seconds 🧵
5
4
2
74
7,381
Nusret retweeted
every ai team ships changes all the time. a new model, a new tool, a prompt edit each one can fix one bug and quietly break another the agent harness makes your model useful. the eval harness tells you what broke before a customer does
31
58
12
671
84,108
open sourced: RL environments for anyone, in any domain.
Introducing rl-env-use: RL environments for anyone, in any domain. Describe a skill in one line. Get a verified environment built from real cases, ready for hillclimbing. 100% open source 🔥
15
48
2
681
40,623
Nusret retweeted
90% of AI Engineer interviews in 2026 are just these 15 concepts repeated: 1/ Transformers & LLM Fundamentals: how LLMs predict the next token, self-attention, multi-head attention, positional encoding, and why transformers replaced RNNs 2/ Tokenization & Embeddings: BPE, how text becomes tokens, and how vectors capture meaning 3/ Prompt Engineering: system prompts, few-shot learning, chain-of-thought, and structured outputs (JSON/schema) 4/ RAG: chunking strategies, retrieval, context injection, context windows, and why RAG beats fine-tuning for fresh data 5/ Vector DBs & Search: vector databases, semantic search, hybrid search , reranking, and HNSW indexes 6/ Function Calling, Tool Calling & MCP: how LLMs trigger APIs, code and databases, and the standard protocol for connecting them 7/ AI Agents & Agentic Workflows: plan, act and observe loops, orchestrated multi-step pipelines, and AI coding agents 8/ Memory & Multi-Agent Systems: short-term and long-term memory, and specialized agents collaborating on a task 9/ Fine-Tuning: full fine-tuning, LoRA, QLoRA, PEFT, and when to fine-tune vs prompt vs RAG 10/ Alignment & Distillation: RLHF, DPO, and knowledge distillation (big model teaches small model) 11/ Quantization & Model Compression: INT8/INT4, pruning, sparsity, and the quality vs speed vs cost tradeoff 12/ Inference & Serving: prefill vs decode, KV cache, speculative decoding, LLM serving, vLLM, batching, and model routing 13/ Multimodal AI & Vision-Language Models: text, image, audio and video in one system, and models that see and reason about images 14/ Evals & Hallucination Detection: LLM evaluation, LLM-as-judge, prompt evaluation, and catching confident but wrong answers 15/ Safety, Observability & Production: guardrails, AI safety, prompt injection, tracing, cost optimization, and production GenAI systems Know these 15 deeply and you can crack most AI engineer interviews
31
111
7
843
50,002
🚀 Introducing Next-Concept Modeling (NCM): our new paradigm for fast, parallel, coarse-to-fine text generation. Compared to strong diffusion language models (DLMs) at similar parameter size, NCM achieves: ⚡ Up to 31× faster generation on downstream tasks like summarization and machine translation with comparable or better quality. 🏎️ 10–21× faster unconditional generation across TinyStories, LM1B, and OpenWebText, with comparable or better generative perplexity and using 3–5× fewer training tokens. 🧩 Better performance on reasoning tasks like Sudoku, outperforming both AR and diffusion baselines with fewer sampling steps. With @ishanskhare, @IddahMlauzi, @ekellbuch, @simran_s_arora, @Azaliamirh, and @HazyResearch. More details in the thread 🧵👇
62
214
62
1,138
503,864
Replying to @mertcobanov
6
4
3
327
27,211
welcome to the 9th annual @stateofai report! ai building better ai, world models, physical ai, geopolitics, cyber defense and 12-month predictions. read it at stateof(dot)ai or watch my editor’s cut here. a few of this year’s biggest findings follow:
47
127
93
704
155,832
Grok Bot could be the first thing you hire instead of prompt It can own a job, work inside your tools for hours and come back with finished work In this article, I show you how
163
622
149
4,521
65,321,153
Yani çok iyi çocuksun ama ben ilişki düşünmüyorum ben birini hayatımda istemiyorum dediğiniz erkekler mi
Dümdüz erkek yok mu işinde gücünde spor yapan piyasacı olmayan sadık kalabilen
86
335
56
16,261
437,640
Nusret retweeted
We’re releasing Olmo-core 3—open training infrastructure for large mixture-of-experts (MoE) models. It’s a core system behind the next generation of Olmo, designed to scale into the trillion-parameter range. 🧵 💻 GitHub repo: github.com/allenai/olmo-core
16
99
13
711
36,747
‼️The Bitter Lesson for context management: Giving LMs unrestricted control over their context beats human-designed SOTA! Introducing 🩵Context Language Models (CLMs)🩵 - Natively manage their own context - Treat context as a file - Learn policies in CLM weights, no harness
102
281
102
2,136
456,195
Bir yıllık model inference loglarını içeren bir veriseti paylaşıldı. 6.12 milyar istek ve 9 bin farklı modele ait veri içeriyor. Link: data.agentic-system.org/
2
42
2,793