Search local San Francisco info

Neighborhood-focused results

4SanFrancisco is a search engine built specifically for San Francisco. We combine multiple indexes, local expertise, and AI systems to rank and surface results that matter to people who live in, work in, or visit the city. Use the site to find businesses, local pages, neighborhood guides, maps, events, and timely news that general search may undervalue. Part of the 4SEARCH network of topic specific search engines.

In Game 4 against the Atlanta Dream, the Golden State Valkyries rediscovered their grit to tie the WNBA Finals at 2-2

The 2026 WNBA Finals are knotted at 2-2 after the Golden State Valkyries won Game 4 on Saturday afternoon in State Farm Arena, defeating the Atlanta Dream, 67-62. As indicated by the final score, Game 4, like Game 1, was played on the Valkyries' terms. However, similar to Game 1, the Dream were right there, with a legitimate chance to escape with the win in the waning minutes, despite an overall off afternoon. With just over 90 seconds remaining, it appeared that Angel Reese, after a rough outing, just might emerge as the hero, as a Reese layup cut the Golden State margin, which has swelled to as many as 16 points, down to five points. After a Dream defensive stop, Atlanta had two chances to slice the lead to two points, as both Allisha Gray and Izzy Borlase missed good looks from 3. Veronica Burton then, once again, came through in the clutch for Golden State, hitting a midranger give…

974 words6 outlets
More in Pro Sports
ajc.com

Atlanta Falcons can beat the San Francisco 49ers on Sunday. Here’s how.

FLOWERY BRANCH — It’s only been a few weeks since Bijan Robinson last talked to Raheem Morris, a figure the Falcons’ star running back describes as a “big brother.” For this week, however, Morris is an evil stepsibling, the 49ers’ defensive coordinator trying to slow down Robinson and the Falcons’ offense. And he’ll do it with an innate knowledge of Atlanta’s offensive personnel. After all, Morris coached eight of the Falcons’ 11 offensive starters last season. At 1 p.m. Sunday, he’ll make his return to Mercedes-Benz Stadium for the first time since the Falcons fired him as their head coach in January after a pair of 8-9 seasons.

1,202 words2 outlets
More in San Francisco 49ers

America’s Companies Keep Flirting With Security Nightmare That’s ‘100 Times’ Worse Than TikTok

A rising San Francisco human resources startup’s use of a blacklisted Chinese AI lab’s model is setting off alarm bells as Beijing continues to reel in American companies. Rippling touts itself as a one-stop-shop AI platform for human resources, information technology and finance that services over 20,000 U.S. companies, per its website. The model’s maker, Z.ai, also known as Zhipu, had already made it on to a U.S. government entity list months earlier over its ties to the Chinese military. Open models “can be tampered with in ways that are very difficult to detect, and we haven’t seen a huge number of attacks like this yet. But I do worry that the potential is certainly there,” Fedusiak warned. Rippling is not the only company to take Chinese AI for a test drive. Giants like DoorDash and small startups are increasingly adopting Chinese AI models because they provide nearly-frontier…

1,288 words2 outlets
More in Employers and Institutions

VS Code Can Now Run Claude and Codex in a Separate Agent Host. What That Process Does

VS Code 1.129, released on July 15, 2026, introduced the fix: the Agent Host, "a dedicated process that runs agent harnesses such as Copilot, Claude, and Codex." Two September releases then let that process run somewhere else. 1.138 put agent sessions inside a local Dev Container, and 1.139 extended that to SSH, Tunnel and WSL hosts. What you give up when an editor renders an agent's session, instead of running the agent's own command-line interface, is the real story. The opposite design exists too, and I picked it for my own tool. The Agent Host docs are short and precise: "The Agent Host is a dedicated process that runs agent harnesses and manages their sessions." A harness is the agent runtime itself. Per the architecture post, the Copilot harness is built on the GitHub Copilot SDK, and the Claude harness loads Anthropic's Claude Agent SDK, with an adapter that maps sessions, tools…

828 words

Is GPT 2XL already clean? — LessWrong

In which I investigate if Opus is able to interpret neurons better than GPT 4. TLDR: Opus 5.5 is able to identify 3-19% of the neurons of GPT 2-XL based on activations. This is a big improvement over the 0.3% that GPT-4 was able…

949 words
More in OpenAI, Anthropic and the AI Boom