I put Lenovo’s RTX Spark laptop through a completely bonkers AI coding test and it passed

Forget ordinary laptop benchmarks - I tested the $4,000 Lenovo RTX Spark with real local AI coding and a brutal battery ...

Google’s new Android XR tools let developers test apps without needing to buy expensive smart glasses

At the 2026 Android Dev Summit, Google has announced some interesting features coming to both Android XR and Android CLI. Developers can now build and test apps for smart glasses without owning the ...
xenospectrum

Anthropic Drops Coding Interviews on One Team as Claude Outpaces Its Hiring Tests

Anthropic's Felix Rieseberg has removed coding interviews from his team's hiring process. On the performance team, the second ...

Anthropic engineer declares the coding interview dead

Coding interviews at Microsoft, and later Google, Amazon, Facebook, and other tech giants, vary, depending on what the hiring ...

An Anthropic team has stopped doing coding interviews: It's not the score that measures a person, but the reasoning behind their decisions

Hello! I'm Uniko🦄!Felix Rieseberg, an engineering manager at Anthropic, announced that his team has stopped doing coding ...

Anthropic manager says the 'coding interview is dead' — here's his advice for junior developers

Felix Rieseberg says AI has made coding tests obsolete. He urged junior engineers to instead show business impact through ...

GitHub Copilot Fusion mixes models to achieve near-Opus 5.5 coding performance at half the per-task cost

GitHub Chief Product Officer Mario Rodriguez confirmed support for continuing work across devices, including mobile. That ...
ctol.digital

Cognition's tougher coding test: a 45.5% top score and a costly final 2.1 points

Cognition's tougher FrontierCode 2.0 lowers the top score from 54.6% to 45.5%. Moving Opus 5.5 from xhigh to max effort buys ...

All tests passed. But this code still cannot be merged

All tests are passing. There are no errors.When I see a row of green checks on the PR screen, I feel relieved for a ...

Microsoft’s New Tool Can Cut AI Coding Costs by Up to 31%

Researchers from Microsoft and the University of California, Berkeley, have introduced a free tool called TokenEconomizer ...

Mistral Large 4 'Le Chonk' arrives as the top open model outside China with improved coding performance and 63% lower cost-per-task

It remains six points behind the open-weight leaders, DeepSeek V4.1 Flash 1017 and GLM-5.3, both tested at their maximum reasoning settings.
Tribune Online

Vibe coding: How non-developers are building apps with AI, where it breaks

Lovable hit $100M ARR in 8 months. Replit's own AI agent also deleted a production database. An honest look at what vibe ...