Urgent.News

What's breaking now, across thousands of outlets.

Tech

What governments put out to tender in software, IT and marketing: November 2026 in numbers

Public buyers in the UK, the US, Canada, Ireland and the rest of Europe publish their tenders as open data. I counted the ones for digital work published between 2 and 16 November 2026: software, IT and cyber, websites, data, marketing, video, translation and e-learning. There were 312 of them. Here is what they ask for, who buys, how long you get to answer and how often anyone tells you the…

Between November 2nd and 16th, 2026, governments in the UK, the US, Canada, Ireland and other European countries released a total of 312 tenders related to digital work such as software, IT, cyber, websites, data, marketing, video, translation and e-learning. The United States accounted for 109 of these notices, the UK for 81, and the rest of Europe for 66, with Ireland and Canada contributing 29 and 27 respectively.

Software was the most common type of work requested, accounting for half of all notices (153). IT and cyber services came in second with 79 notices, followed by marketing and creative work (52), and data and AI applications (32). In the US, median deadline was the shortest at 14 days while the UK and Ireland had the longest at 27 and 32 days respectively. Most notices (83%) included a budget, with Irish and UK notices being the most transparent, while none of the US or Canadian notices disclosed a budget.

The US federal procurement system saw the most technical and complex notices, with IT and cyber (46) and software (45) making up almost all of them. Defence agencies such as the Air Force, Navy, Army, and various agencies like DISA and Defense Logistics Agency were the primary buyers. In the UK, the mix of industries was broader, with software leading (39) and marketing and creative work second with 23 notices.

Irish tenders were mostly focused on health, food, education and local government, while Canadian tenders were primarily driven by federal and defence agencies.

Across all notices, the median gap between publication and deadline was 25 days, with a quarter closing within 14 days. The US, however, was the outlier, with a median of only 14 days. European tenders were generally aligned with the EU’s minimum periods, typically around 30 days. In the US, a few "Sources Sought" notices, requesting information rather than bids, were present, with 20 of the 109 US notices being set aside for small businesses.

The largest values in this fortnight included €6bn for EuroHPC’s AI gigafactories, £350m for London’s network services contract, and €27.2m for ESB’s VMware licences.

Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at dev.to →

More in Tech

This e Binding: why the value changes with each call

Introduction This is one of the parts of JavaScript that generates the most confusion, because its value is not always what we expect when reading the code.

  • "this" keyword in JavaScript causes confusion
  • Early binding decided by static facts before execution
  • Late binding decided by dynamic facts during runtime

How to Remove Blank Pages from PDF with Python

Unwanted blank pages can appear when scanning, merging, or exporting PDFs. For example, duplex scanning may capture the empty backs of sheets, while document layouts can generate unnecessary blank…

  • Python script removes blank PDF pages using Spire.PDF and Pillow
  • IsBlank() method and SaveAsImage() detect blank pages
  • Pages are removed in reverse order to avoid index shifting

Simple S3-Compatible Object Storage for Storing User-Uploaded Avatars Safely

The practical answer is to use private, S3-compatible object storage and make retention a policy owned by your application.

  • Infrai provides S3-compatible object storage for user avatars
  • Monitor deletion candidates by tenant and object restoration metrics
  • Compare deletion count to policy threshold for effective guardrails

Qwen3-235B-A22B: China's Case for Large-Scale Open-Weight Inference

Originally published on juliancardozo.github.io. Qwen3-235B-A22B-Instruct-2507 condenses a trend that is redefining the open-weight market: large capacity models that activate a fraction of their…

  • Qwen3-235B-A22B from Alibaba uses only 1% of its 235B parameters per token.
  • Model supports 119 languages and dialects under Apache 2.0 license.
  • Extended context of 262K tokens may increase computational costs.

More from Tuesday 17 November →