The Codebase Map for AI Agents, Teams and Applications. https://xitter.cf/t.co/geL6S8o7Et
Remote
Joined April 2023
- Tweets499
- Following95
- Followers557
- Likes615
We’re excited to announce that Code Fundi is officially leveraging Gemini’s new Multimodal File Search to help autonomous agents reclaim over 50% of their context window.
Read the full @Google blog post to see how we’re using it to power our search API.
blog.google/innovation-and-a…
Code Fundi retweeted
Introducing Dispersl - the world's first AI Dev team.
Dispersl’s AI agents collaborate with each other to build full-stack apps for you, and they can also work for your AI agents.
It also supports MCP servers, letting you plug in any tool.
Comment "DM" to get early access.
Code Fundi retweeted
Over 15 million tokens processed by @code_fundi just this year alone. That’s like half a million tweets worth of code
Crazy that something I built from my bedroom in Kenya got this much traction globally
Try it out in Cursor, Windsurf and VS Code: 👇
marketplace.visualstudio.com…
DeepSeek-R1-0528 now available in CodeFundi!🚀
Start building with DeepSeek's most powerful reasoning model that outperforms Gemini 2.5 Pro and Qwen3 in math, coding, and logic tests.
🚀 DeepSeek-R1-0528 is here!
🔹 Improved benchmark performance
🔹 Enhanced front-end capabilities
🔹 Reduced hallucinations
🔹 Supports JSON output & function calling
✅ Try it now: chat.deepseek.com/
🔌 No change to API usage — docs here: api-docs.deepseek.com/guides…
🔗 Open-source weights: huggingface.co/deepseek-ai/D…
The context window remains 200K tokens, just like Sonnet 3.7.
It doesn’t match Gemini 2.5 Pro’s 1M+ tokens, but Anthropic’s optimizations for context handling keep performance strong.
How have you been using Claude 4 so far and what stands out to you?
Claude Opus 4 targets intricate, multi-step workflows. It can generate and manage “memory files” to preserve state across sessions, much like CodeFundi’s long-term memory feature.
Anthropic has released Claude 4 and it arrives with two variants: Sonnet 4 and Opus 4!
Both excel at code tasks, with Sonnet 4 hitting 72.7% on SWE-bench and Opus 4 close at 72.5%. These results set a new bar for open models.
You can also use your own Anthropic API key to get unlimited usage with the new models!
Try it out in Cursor, Windsurf and VS Code: 👇
marketplace.visualstudio.com…
Claude Opus 4 & Sonnet 4 now available in CodeFundi V0.26!🚀
Try out Anthropic's most powerful coding and agentic models in your workflows today.
👀 Projects can use fresh code or extend existing code
📦 Our Filecoin-powered bounty challenge drops June 5
🔓 Build the systems of the future — secure, sovereign, and open by design.
#PLGenesis @PL__Genesis
⚡ Builders, your moment starts now!
We’re proud to partner on PL_Genesis: Modular Worlds, a new virtual hackathon from @protocollabs & @PL__Genesis
💰 $250K+ in prizes + a path to the Founders’ Forge pre-accelerator
Register now and get ready to build:
bit.ly/45jZ1mz
PL_Genesis: Modular Worlds is live!
🌍 Build the future of the internet—secure, open, and intelligent. $250K+ in prizes, mentorship and more.
Join the movement: plgenesis.com
#PLGenesis #ModularWorlds
