Pieces for Developers

Pieces for Developers

@getpieces.bsky.social

The first AI that remembers everything you do. Long-Term Memory Agent for your favorite LLMs. linktr.ee/getpieces

162 followers55 following572 posts

Introducing "Pieces of Memory". A brand new AI webinar series by builders, for builders. 🛠️ Bi-weekly sessions on: ➡️ Building long-term memory for AI ➡️ Productivity hacks that actually scale ➡️ Latest LLM/agent architectures First episode on July 10th at 9am. Sign up below.

OpenAI solved the intelligence problem. Pieces solves the memory problem. Context from Slack. Context from your standups. Context even when apps crash. This is what it looks like when your copilot actually remembers your workflow.

Are AI agents about to reshape how we build software? In a recent piece, we break down why this shift isn’t just about faster code generation or better autocomplete. The teams that figure out how to work with agents (not just use them) will be the ones that win. Read the full article here 👇

Stop copy-pasting old solutions. With Pieces + Cursor integration, just ask 'How did we fix this last time?' and get instant access to your past solutions. Your entire development history, searchable through natural language.

Clicked through 20 Chrome tabs just to find the blog post you skimmed yesterday? We've been there. That's exactly why we built Pieces. It's like having a photographic memory for every useful thing you've ever saved. But better.

Ever wonder how you actually solve long-term memory for AI? In a recent interview, Ellie Zubrowski sat down with Principal Research Scientist Antreas Antoniou to chat about nano models, transformers, local AI, and privacy. Check out the full interview here & let us know what you think! 👇

We've all been there - explaining the same project context over and over to different AI tools. That ends now. Pieces MCP Server creates a bridge between your development history and AI assistants, all running locally on your machine.

Pieces OS 12.0.0 is live, introducing sign-in for all users. This isn’t just about security. It’s a foundational step toward a more reliable, consistent, and personalized experience across the Pieces ecosystem. Smarter memory, better agent experiences, and tighter integrations are all on the way.

Join us for the Pieces Community Sync - today at 1pm ET. (in about 1 hour!) This is a good chance to connect with the team, ask questions, share feedback, and see what others are building with Pieces. Make sure to set a reminder 👇

Going live in a few hours 🎥 Episode 1 of The Missing Pieces drops today @ 12pm ET. Ellie Zubrowski is chatting with Hanna Stechenko, our Head of SEO, about her path into devtools and how her work helps devs discover Pieces. 🔗👇

🚨 New series alert 🚨 Meet the people building Pieces 👋 The Missing Pieces premieres tomorrow - June 4 @ 12pm ET. First up: Ellie Zubrowski sits down with Hanna Stechenko, our Head of SEO, to talk about her career path and how devs end up finding Pieces in the first place.

Not all AI models need to be massive. As Sam Jones (Chief AI Officer at Pieces) put it: “I think 90% of people working in software, if you tell them this does the same job, it’s cheaper, it’s faster, and you don’t have to ping an API… I think everyone wants that.”

Rewind AI isn’t what it used to be… and honestly, it never really worked for developers. If you’ve been thinking about switching, this post walks through the best memory-first tools in 2025 - from privacy-safe screen recorders to code-aware long-term memory agents.

In LTM‑2.5, Workstream Activity got a serious glow-up 👀 Temporal groupings, improved navigation, related Copilot chats - a whole new hierarchy in the UI to keep you oriented and in flow. More context, less confusion. Have you tried the new Workstream Activity view? We'd love your feedback!