Nove modelli frontier messi alla prova sulla ricerca web più difficile mai testata 🔎 DeepWeb-Bench richiede di raccogliere prove massicce da fonti diverse e derivazioni a lungo raggio, molto più complesse dei benchmark esistenti. Dentro troverete: 📌 Solo il 12-14% degli errori dipende dal retrieval, il resto da derivazione e calibrazione 📌 Modelli forti e deboli sbagliano in modi qualitativamente diversi 📌 Accordo …

Channel
Datapizza Community 🍕📈
@datapizza
On this record: Growth · Engagement · What this channel posts · Reactions · Posts · Citations · Cite this entry
14,773subscribers
-18 since we began measuring on 6 August 2026
Risers and fallers across the register · movement among entries of 10,000–31,623.
Register entry
| Telegram ID | -1001308702422 |
|---|---|
| Type | Channel |
| Username | @datapizza |
| Description | Canale Telegram ufficiale della Datapizza Community📍 👉🏾 https://datapizza.tech/ Contiene fonti, approfondimenti, extra e spunti di discussione sul mondo Tech & AI👩💻👨💻 |
| Created | Between 1 March 2018 and 31 July 2021— estimated from Telegram’s id allocation, not measured. How this range is calculated. |
| First recorded | 6 August 2026 |
| Last confirmed live | 14 August 2026 |
| Measurements held | 8 |
| Confirmed unchanged | 1 time, most recently 14 August 2026 |
| On Telegram | t.me/datapizza |
Growth
| Measured (UTC) | Subscribers | Change |
|---|---|---|
| 14 Aug 2026, 07:24 | 14,773 | -4 |
| 12 Aug 2026, 23:35 | 14,777 | +1 |
| 11 Aug 2026, 23:27 | 14,776 | -3 |
| 11 Aug 2026, 02:51 | 14,779 | -2 |
| 10 Aug 2026, 02:02 | 14,781 | -3 |
| 9 Aug 2026, 05:02 | 14,784 | -7 |
| 7 Aug 2026, 02:14 | 14,791 | no change |
| 6 Aug 2026, 18:18 | 14,791 | first reading |
Engagement
24 posts held, back to 13 July 2026 — the reader has not yet reached the start of this channel’s public history, so older posts may sit further back, unread. Read across 20 pagesof Telegram’s post history, 20 posts per page.
- ERR · 30 days
- 21.8%
- avg views ÷ 14,773 subscribers
- Avg views / post
- 3,220
- 20 posts measured
- Reaction rate
- 0.159%
- reactions ÷ views · ER floor
- Posts in window
- 20
- of 24 held
ERR is average views per post over the last 30 days divided by subscribers, the definition TGStat uses, so this figure is comparable with the one you will see elsewhere. It falls structurally as a channel grows: a high ERR on a small channel and a low one on a large channel describe reach mathematics, not quality. We publish the figure and the sample it came from and pass no verdict on it.
ER is defined industry-wide as (forwards + reactions + comments) ÷ views— note the denominator is views, not subscribers. Telegram’s public web preview carries views and reactions but not forward or comment counts, so the reaction rate above is the reactions term only and is therefore a floor: the true ER for this channel is higher by an amount we have not measured and will not estimate. It is computed over the 18 of 20 measured posts that carry a reaction reading, and over those same posts' views.
| Window | Rolling 30 days · latest post in window 14 August 2026 |
|---|---|
| Posts held | 24 (13 July 2026 – 14 August 2026) |
| Views total | 64,450 |
| Reactions total | 96 |
| Forwards / comments | not exposed by the public surface — not measured, not estimated |
| Readings taken | 15 Aug 2026, 08:41 UTC |
Views are a single reading per post, taken at the time above. A post published in the last day or two is still accumulating views, which pulls the 30-day average down slightly. That is a property of the standard definition rather than a fault in it, so we keep the definition rather than “correcting” the number into something nobody can reproduce.
Precision. Telegram publishes view counts on its public widget in short form — 8.12K, 3.7M — so any reading at or above 1,000 reaches us rounded to three significant figures, and only counts below 1,000 are exact. Averages and rates derived from them are shown to the same precision rather than to the unit: a figure like 3,701,250 would assert digits nobody measured.
Reaction counts are published per emoji and rounded the same way, so a total below 1,000 is exact and a larger one is a sum that may carry a rounded component from each emoji above 1,000. Because it is a sum, it does not look rounded — read a large reaction total as three significant figures per contributing emoji rather than as the figure it prints.
What this channel posts
- Photos
- ≈730
- Videos
- ≈19
- Links
- ≈1,000
Lifetime counters from Telegram’s own channel header, read 15 August 2026 — not the date at the top of this page, which is when the subscriber count was last read. A count marked ≈ was rounded by Telegram before we ever saw it — t.me prints these counters in full below 1,000 and to three significant figures above, so ≈142,000 means somewhere between 141,500 and 142,499.
Reaction mix
106 reactions across 21 posts, in 8 distinct kinds. The most used accounts for 79.2% of them.
| Reaction | Count | Share | Share, drawn |
|---|---|---|---|
| ❤ | 84 | 79.2% | |
| 🔥 | 11 | 10.4% | |
| 👍 | 4 | 3.77% | |
| 💯 | 3 | 2.83% | |
| ❤🔥 | 1 | 0.943% | |
| 🐳 | 1 | 0.943% | |
| 👏 | 1 | 0.943% | |
| 🤣 | 1 | 0.943% |
No sentiment is inferred, and none should be read in. This table is ordered by count and by nothing else. Emoji do not carry stable meaning across languages or communities — 🙏 is thanks in one channel and mourning in another — so we publish which ones were pressed and how often, and pass no judgement on what an audience meant by them.
Precision. Telegram publishes reaction counts per emoji and short-forms each one — 4.34K, 1.2M — so any single kind at or above 1,000 reaches us at three significant figures, and only counts below 1,000 are exact. The shares above are ratios of those figures and carry the same error. This is also why the total here can differ slightly from a reaction total printed elsewhere on the page: both are sums of the same rounded parts, taken over samples with different edges.
Coverage. Reactions were read on 22 of the 24 sampled posts in this sample. Summed by Telegram’s own count on each post — not by adding up the per-emoji breakdown above — those same posts carry 106reactions in total: the kind of figure the paragraph above means by “a reaction total printed elsewhere on the page”.
Measured over the 24 most recent posts we hold, published 13 July 2026 to 14 August 2026, using the newest reading held for each. Telegram Stars are excluded: they are a payment, not a reaction, and they have their own section.
Recent posts
Se in questo periodo state cercando nuove opportunità, su Datapizza Jobs abbiamo due posizioni che potrebbero interessarvi! 🙌 1️⃣ Product Lead in Medicilio 💸 RAL: 50k-75k 💡 Esperienza: 4+ anni 📍 Ibrido su Milano (3 giorni smart-working/settimana) 🙌 Tecnologie: UX/UI Design, AI Fluency, Comunicazione efficace, Product Strategy, Microservizi, Gestione Team, Project management, Regulatory Compliance, SQL 👉 https://bit.…
❤5
Un framework per testare le vostre applicazioni LLM come fareste con del codice normale 🧪 DeepEval è una libreria open source con decine di metriche pronte per agenti, RAG, chatbot e summarization, con una developer experience in stile Pytest. Dentro troverete: 📌 Metriche: agenti, RAG, chatbot, summarization già pronte all'uso 📌 Integrazione: si collega ai principali framework LLM 📌 Licenza: Apache 2.0, completamen…
❤10
Un dataset multilingue da mille miliardi di token è online 🌍 Hugging Face ha rilasciato FineTranslations: un dataset di testo parallelo che copre l'inglese e oltre 500 altre lingue, pensato per migliorare la traduzione automatica a minori risorse. Dentro troverete: 📌 La scala: oltre 1000 miliardi di token di testo parallelo 📌 La fonte: contenuti non in inglese di FineWeb2 tradotti con Gemma3 27B 📌 La pipeline: proc…
❤9
Un paper per ripensare la memoria degli agenti AI 🤔 I ricercatori propongono Memory in the Loop: un sistema di retrieval in-process che funge da memoria di lavoro estesa per gli agenti linguistici, senza ricorrere a database esterni separati. Dentro troverete: 📌 Il problema: la memoria a lungo termine tradizionale rallenta gli agenti 📌 La soluzione: retrieval integrato direttamente nel processo di inferenza 📌 I van…
❤2👍1
Se in questo periodo state cercando nuove opportunità, su Datapizza Jobs abbiamo due posizioni che potrebbero interessarvi! 🙌 1️⃣ Lead Data Architect in Intarget Group S.r.l. - Società Benefit 💸 RAL: 60k-70k 💡 Esperienza: 5+ anni 📍 Ibrido a Milano, Pisa o Roma 🙌 Tecnologie: Python, SQL, Apache Kafka, GCP, Data Pipeline, Strumenti BI, CI/CD, Terraform 👉 https://bit.ly/4giJbO4 2️⃣ AI Transformation Leader in AzzurroD…
❤6
Portare l'AI in azienda sembra facile. Poi provate a farlo davvero e scoprite quante cose si possono rompere. Un anno di Adoption ci ha insegnato dove serve attenzione, persone, processi, tempi. E cosa serve perché il cambiamento resti in piedi. In questa live troverete: 📌 Casi reali, 📌 Cose imparate sul campo 📌 Nessuna teoria da slide E se vi siete mai chiesti com'è lavorare su questo lato di Datapizza, qui potr…
❤4
Un agente che decide da solo come strutturare il proprio ragionamento 🧠 I ricercatori dell'University of Washington propongono Deep Reasoning, un metodo per costruire scaffold di ragionamento su misura per ogni task, invece di usare pattern fissi. Dentro troverete: 📌 Un linguaggio formale per rappresentare il meta-ragionamento 📌 Combinazione di inferenza associativa, calcolo formale e scomposizione ricorsiva 📌 DOLO…
❤4🔥2
Le AI imparano davvero dai propri errori? ❌ Epoch AI ha lanciato EBR-bench: fa giocare ai modelli frontier lo stesso gioco da tavolo complesso, Earthborne Rangers, per 30 partite di fila. Il risultato mette in discussione una delle promesse più ripetute sull'apprendimento continuo. Dentro troverete: 📌 Setup: modelli valutati su 30 playthrough consecutivi dello stesso gioco 📌 Risultato: nessun segno di miglioramento…
Un benchmark per capire se gli agenti sanno davvero fare code review 🦀 c-CRAB è il nuovo dataset pensato per valutare quanto gli agenti AI siano affidabili quando devono analizzare pull request reali, non solo scrivere codice da zero. Dentro troverete: 📌 Task: revisione di pull request reali su repository di produzione 📌 Metriche: precisione nel trovare bug, stile e problemi di design 📌 Confronto: gap ancora ampio …
❤4
Quanto è cambiata l’AI dal 2023 ad oggi? 🤔 Nel 2023 gli agenti AI completavano da soli compiti di pochi minuti. Oggi si avvicinano alle 12 ore! ⌛️ Un esempio concreto: "organizzami la giornata di domani". Nel 2023 ricevevate una lista di consigli generici. Oggi un agente legge il calendario, le email, nota i conflitti, riorganizza le priorità e vi presenta un piano pronto da approvare (e senza il bisogno di saper s…
❤6
Vediamo insieme le nuove scoperte di Anthropic che per l'ennesima volta ci fanno ricredere su quanto credevamo di sapere. 📌 Cosa sappiamo davvero su perché funzionano gli LLM? 📌 Le nuove evidenze suggeriscono che i modelli possono pensare ? 📌 E quindi sono coscienti? (no, ma capiamo insieme il perché) Ne parliamo in live con: Manuel Vimercati, Direttore Scientifico Irene Senatore, AI Adoption Specialist 🗓️ Giovedì…
🔥3
Showing the 12 most recent of 24 posts we hold for @datapizza. View and reaction counts are the latest single reading for each post, not a live figure, and a recent post is still accumulating both. A view count marked ≈ was rounded by Telegram before we ever saw it — t.me prints views in full below 1,000 and to three significant figures above, so ≈1,200,000 means somewhere between 1,150,000 and 1,249,999. Unmarked counts are exact. Text is reproduced from the public post preview and truncated for length.
Citation-graph rank
Citation-graph rank — 802,674 of 1,478,351entries in the measured graph. A weighted position computed from the forward and mention edges below — republished posts weigh more than named mentions — and recomputed periodically, over the whole graph. Published only as this ordinal position, never as a score: a position is a fact, and a score printed beside one channel’s name would read as a verdict this register does not make. The two counts beneath stay separate for the same reason mentions are never summed with forwards anywhere else on this page — a named-by count costs nothing to manufacture. The top 100 by this measure, or how it is computed.
Forward network
Republished by
Channels on the register that have forwarded this channel's posts into their own feed.
Built only from forwarded posts we have actually read, on both sides. Coverage is early and deliberately incomplete: a missing link means we have not read the post that would prove it, never that the relationship does not exist. Counts are distinct forwarded posts observed, so they only ever go up as we read more.
Cite this entry
A live page changes as we take new readings, so a citation should name the measurement it is based on, not just the URL. The line below cites the subscriber count as measured 14 August 2026 — this entry's latest reading, not the date you are reading this.
“Datapizza Community 🍕📈” (@datapizza), 14,773 subscribers as measured 14 August 2026. Telegram Register, tgregister.com/channel/datapizza.
Full measurement history, CC BY 4.0. Every reading this register holds for this entry, not just the latest one, as a dated, downloadable record: CSV · JSON. Free to use with attribution to tgregister.com. Each file carries its own generation timestamp, which is the figure to cite for exactly when the data was retrieved.