PWV Portfolio Feed

“Make it more minimal.” → “Maybe not that minimal.”

✨ New in Agent Runners✨: Revert to a previous run.

For when you iterate a just little too far.

Because you want the code to go through the exact same processing as in your main Vite app

Tech layers are fluid. Abstractions like Fastlane or generated projects filled gaps, but as layers evolve, some become unnecessary.

At Tuist, automation is LLMs + bash now. We're moving to vanilla Xcode as things mature. GitHub's "runs-on" signaled a similar shift.

Balance sales with product.

monthly allowances trailing core inflation?

At Tuist, we abstract only as a last resort. We stay close to Apple's approach—when solutions emerge at the right layer, we integrate and extend rather than abstract.

That's why we're excited about caching capabilities emerging at the right layer, though not yet as robust as module-based caching. 🚀

Zed Pro users now have access to AnthropicAI's Claude Haiku 4.5 (no app update required). For users running through Anthropic or AWS Bedrock providers, Haiku 4.5 is available in Preview (Zed 0.210.0+) now, and will be in stable next week!

Thousands of devs have already used Netlify’s Agent Runners to speed up their workflow.

Top use cases: 🧩 build fixes (32%), 💬 chat prompts (24%), 🎨 layout edits (20%).

Ask Mode’s coming next. What else should Agent Runners handle? 👀

same, and i'm mostly using neogit in neovim and there's no jj alternative yet. but i am really optimistic about a simplified scm and what a purely jj-based system could do

Additionally, Elicit will add your keyword query to the Report created from your Systematic Review. Anyone who reads your Report can go back and quickly reproduce the results.

Elicit will show your query history within a Systematic Review as well. 

Then, you can revisit keyword queries, compare their results to semantic search queries, and iterate until you feel confident that your query captures the right set of papers.

In many Systematic Reviews, researchers need to limit their search to one corpus at a time.

Now, you can choose to search over all Elicit papers (including PubMed), only over PubMed, or only over ClinicalTrials(.gov). Our keyword queries will auto-translate for each corpus too.

Transforming a PICO question into a solid keyword query can be time-consuming and research-intensive.

To help solve this, Elicit automatically turns your plain-language research question into a polished draft keyword query, complete with synonyms and MeSH terms.