Apps by Prompt, Video by AI, Front-End by Cursor

— plus warm-intro jobs inside

In partnership with

Friends!

Short and sweet: big product shifts this week; the jobs (with warm intros) are at the bottom. ⬇️

The Simplest Way To Create and Launch AI Agents

Imagine if ChatGPT and Zapier had a baby. That's Lindy.

With Lindy, you can build AI agents in minutes to automate workflows, save time, and grow your business. From inbound lead qualification to outbound sales outreach and web scraping agents, Lindy has hundreds of AI agents that are ready to work for you 24/7/365.

Stop doing repetitive tasks manually. Let Lindy's agents handle customer support, data entry, lead enrichment, appointment scheduling, and more while you focus on what matters most - growing your business.

Join thousands of businesses already saving hours every week with intelligent automation that actually works.

( If you don’t want ads like these, Premium is the solution. )

Lovable AI+Cloud: full-stack apps by prompt (Replit Agent 3 vibes)

Lovable just pushed its “AI+Cloud” story hard: chat to build, deploy, and wire up backend pieces. You focus on intent; it scaffolds auth, DB, storage, file uploads, and deploys — then you refine.

In practice, you’ll still connect providers for serious AI features, but Lovable removes a ton of setup friction. (Lovable Documentation)

Lovable claims >100k new ideas/tools/sites per day; the company spun out of the GPT-Engineer project in 2023.

Quick wins you can prompt today

  • “Add email/password login and link entries to users.” (Supabase + auth wired.) (Lovable Documentation)

  • “Add profile picture upload; store in storage and update profile.” (Edge function + validation.) (Lovable Documentation)

  • “Connect payments with Stripe; gate features by tier.” (Uses webhooks + roles.) (Lovable Documentation)

  • Sabrine’s Plinq (women’s safety) — impressive traction story.

  • Lumoo’s AI content platform — EU startup profile.

  • Rory’s marketplace — founder page & build notes.

Bottom line: Lovable AI+Cloud is the closest “no excuses” path from prompt → live app right now. I’d prototype here, then graduate hot projects to your main stack as needed.

Sora 2 lands: better physics, control & realism — and sharper risks

OpenAI’s Sora 2 is out alongside the Sora app for creator-grade shorts with synced audio. The jump from Sora 1 looks real on physical consistency and control.
Top positive: pro-level video creation gets democratized.
Top negative: deepfakes get harder to spot — policy, provenance, and detection must keep up.

The launch video:

And this one, ahaha:

Cursor + Claude + Figma MCP: the front-end gap keeps shrinking

Design-to-code is growing teeth. Figma’s MCP server feeds real design data into agents; Cursor and Claude Code both ride this, and Cursor now exposes an Agent “planning” capability (structured to-dos) so you can plan → execute cleanly.

Model tier picks (opinionated):

Pair it like this: plan with GPT-5-Codex, execute with Sonnet 4.5 — fast, cheap, but thoughtful planning; relentless implementer.

And yes, GPT-5-Codex is a thing — OpenAI’s agentic-coding variant — live in Cursor and documented by OpenAI. If you write software, you should try it this week.

Cursor also has a native planning mode:

The image is a screenshot of a coding assistant interface, showing a conversation about improving code maintainability across two repositories named aibpo-backend and aibpo-frontend-v2. At the top, a yellow command box displays: /improve_code. Below, the assistant writes a detailed message with four numbered clarification questions: Whether the focus should be on the backend, frontend, or both. Which parts of the system are the most painful (routes, services, data access, UI state management) and should be prioritized. Whether proposals should be limited to incremental, non-breaking refactors, or if large structural changes are acceptable. Whether there are tech-debt items or architectural decisions that should be treated as off-limits (examples given: ORM, queueing system, state management libraries). The user has replied with short numbered answers beneath: “more frontend” “I don't know” “smaller changes better than larger” “i don't know, you tell me” After that, the assistant interface shows a prepared plan card. The card heading is frontend-2b1af009.plan.md followed by a bold title: Frontend Maintainability Review. The description below reads: “Assess the frontend codebase to surface the top three maintainability improvements that preserve current behaviour while simplifying structure.” A blue link labeled “Read detailed plan” appears beneath. At the bottom right corner of the plan card is a yellow button labeled “Build ⌘⏎”. The overall design of the interface is clean, with light background, gray boxes for messages, and subtle color highlights (yellow for commands and build button, blue for links).

Our Weekly Jobs 👨‍🏫

Subscribe to Premium for more warm referrals

Subscribe to Premium to See the Rest

Upgrade to Premium for exclusive demos, valuable insights, and an ad-free experience!

Already a paying subscriber? Sign In.

A subscription gets you:

  • • ✅ Full access to 100% of all content.
  • • ✅ Exclusive DEMOs, reports, and other premium content.
  • • ✅ Ad-free experience.