Pocket is shutting down - by now, we all know that. I wasn t a regular user myself, but I did test the app about a year ago.
From what I m seeing in the comments, long-time users are understandably panicking. There are plenty of alternatives out there, but it seems like people are struggling to find something they can easily switch to.
So I m curious - how did you use Pocket?
And have you found a replacement that actually works for you?
Some tools just feel more reliable even if the backend models are similar. Is it the tone, layout, citations, or transparency of the process? What gives you confidence to act on what AI says?
Just yesterday I prevented my team from adding an exotic feature to our product.
My hypothesis is that people don't like many features in a product as that complicates the product adoption e.g. many sales guys hate CRMs for this reason. In that sense, more features might equate to no features as users don't adopt/use the product. So, minimalistic products that solve 1 big problem (80% of the problem pie) is what people like.
Linear for Agents is now GA. Who is building or using an agent for Linear?
We recently built one for using Bucket feature flags within Linear (and shared some of our lessons learned). Would love to hear from others building or using agents. We're just getting started with our Linear integration and are keen to get feedback. We've plans to add:
Feedback/adoption result summaries to the agent
Support for an access summary embed whenever a link to a feature on Bucket is dropped
Plus, integrate our feedback function with Linear's customer requests - making it possible to capture customer requests directly from within your application
One of the joys of vibe coding is seeing an idea take shape rapidly. But how often does that "shape" end up being a frontend-only prototype, and the dream of a fully functional app with your preferred tech stack (say, Next.js + PostgreSQL, or React + FastAPI) feels like a whole separate, slower project?
At Proofs, we're exploring how AI agents can bridge this gap, generating not just UIs but the entire underlying application structure, including backend logic and database setup, using popular open-source stacks.
We're seeing growing interest in our product people visit, explore the homepage, even watch the demo. But most stop short of signing up or engaging with the core features.
We re considering open-sourcing the core product and keeping advanced capabilities (like customization, automation, database support) as paid upgrades. The idea is to remove the try before trust barrier and build credibility through transparency.
We ve seen some platforms go this route and it seems to work. But we re curious:
We re excited to announce two powerful new features in PulpMiner that make API creation from webpages even more flexible and accurate. Check them out below
Like customized AI job searching, AI mock interviews, experience sharing, Gmail and calendar tracking. For me, I need to go indeed to find jobs, use simplify to apply it, interview with ChatGPT to practice, find the real interview questions on Reddit. And manage the job applications by myself to check the email everyday, and the amount of job I applied already hit 400 this month. I m tired.
Just got back from MAU Vegas 2025, where I spent a few days nerding out on GTM tech stacks with folks from consumer apps, gaming, fintech, and lifecycle platforms. If you're building or scaling a mobile-first app, here s the distilled version of what top-performing teams are actually doing right now minus the vendor hype.
What s essential in 2025
The consensus was pretty clear: besides AI everything, the modern GTM tech stack for mobile apps boils down to five key components:
Hey PH fam I launched Lumoar (B2B SaaS startup) three weeks ago and we have already seen over 100 users sign up and lots of positive feedback. It s clear there is a demand and I feel there s potential in this. But now I have hit a wall. Because of my country, international payment platforms like Stripe or PayPal don t provide API access so I can t monetize or implement a paid plan right now. I ve talked to some investors, but they expect initial capital or revenue before they d consider funding. Bootstrapping got me this far, but moving forward without monetization or investment is getting really difficult. If you ve dealt with something like this (or just have thoughts), I d love to hear:
What would you do in this situation?
Any creative approaches, funding workarounds, or even alternative payment methods I should look into? Appreciate any advice
I'm a founder in the depths of pivot hell. But I have a strong feeling that the important space to build in is memory. For me, the thing that can increase the value we generate from AI 10x from where the models currently are, even without further improvement, is memory and context. Being able to transfer context from one interaction to the other, preferences for how responses should be structured, permission management, sensitive information filtering and redacting, etc. Of course, all these have to be done securely and end-to-end. I've thought a lot about how to find this wedge. A lot of things I've found just sound cool but probably aren't providing much value to the user. I'm optimizing for a high-value, high-frequency use case. This is the best-case scenario. Especially for a consumer-based solution. I would list some of the solutions that I think can stem from having an end-to-end encrypted memory of all AI interactions (limited to conversations with AI models for now) and a system that does a basic filtering of sensitive information and PIL:
Smart search: being able to semantically search through old conversation history to find working solutions with pure conversation.
Prompt injection: Being able to use old verify prompts with a click to ask better questions.
Conversation summaries: Just as it implies, get summaries of conversations, perhaps of just one chat or multiple. And maybe even across multiple AI tools.
Project workspaces: Having an organization share a connected memory. With each team member being able to use the other members' context on what they're working on to make sure they're all aligned on the goal and keep a coherent implementation of the tasks.
Chat with memory: Using a lightweight LLM to converse with your chat history across all AI use and different conversations and sessions.
One interesting thing I came across this week was that the CEO of Duolingo first declared intentions to use AI to replace contract workers in some positions. However, they later withdrew that comment, making it clear that AI will not replace its employees.
Ahh, this type of discrepancy appears to be happening more often, to be honest. The same thing happened to Klarna not long ago. That AI will take care of everything in one minute, and then, hold on: in reality, we still need human workers.
@Paste just launched their latest version, and tbh I'm considering switching over but I'm curious to hear what other people use? I know @Raycast has a pretty powerful and integrated one that I'm considering using. I currently use @ Pasta which I love. Do folks have any they recommend? What's the consensus?
