@qdrddr.bsky.social
36 followers 60 following 210 posts
https://targeterpro.com/linktr-ee-qdrddr
Posts Media Videos Starter Packs
📊 RaBitQ by #LanceDB cuts storage up to 32× with only ~5% recall loss — boosting speed & efficiency for #AI workloads.
🔗 Link in first 💬⤵️

Repost 🔁 #RAG #VectorDB #VectorStore #SemanticSearch
🦀 #LanceDB #Graph — embeddable #Cypher query engine in Rust with Python bindings.
Apache 2.0 lic, lightweight #KuzuDB alternative.
🔗 Link in first 💬⤵️

Repost 🔁 #GraphDB #AI #Rust
🧩 @surrealdb.com — embeddable in-progress #LPG #GraphDB, #SQL & #JSON docs, just like #SQLite for your #AI app perfect replacement for archived #KuzuDB
Supports #GraphQL, #VectorSearch, Full Text.
Python, Rust, JavaScript, .NET bindings.

🔗 Link in first 💬⤵️
Repost 🔁 #Databases #VectorDB #GraphRAG
#Apple: The Illusion of Thinking:
Understanding the Strengths and Limitations of #Reasoning Models
via the Lens of Problem Complexity

ml-site.cdn-apple.com/papers/the-i...
#AI #LLM
ml-site.cdn-apple.com
👉 Discuss with me in Discord: linktr.ee/qdrddr

DuckDB LPG duckpgq.org/documentatio...
Archived Kuzu GitHub repo github.com/kuzudb/kuzu
🗂️ #KuzuDB project archived 2025 🪦— once a fast, embeddable #LPG #graph DB (MIT) with #Cypher & partial #RDF support.
Possible successor: @duckdb.org + DuckPGQ 🦆
Who can fill this gap better? Let me know!

🔗 Link in first 💬⤵️

Repost 🔁 #Databases #GraphDB #OpenSource
#PostgreSQL query pipelining — up to 71× faster under high latency in v18! #Python & #C #C++ supported; Java, C#/.NET still missing
🔗 Link in first 💬⤵️

Repost 🔁 #Databases #Tech #Dev #DevOps #SQL
Reposted
I’m getting big “artists at a con circulating the same $20 around” vibes but like, in an evil way
🛠️ #Cloudflare turns #MCP tools into TypeScript #API — enabling #LLMs to generate & chain code efficiently with fewer tokens
🔗 Link in first 💬⤵️

Repost 🔁 #AI
Code Mode: the better way to use MCP
It turns out we've all been using MCP wrong. Most agents today use MCP by exposing the
blog.cloudflare.com
🤖 #Microsoft unifies #AutoGen & #SemanticKernel into the new #AgentFramework — one ecosystem for building & orchestrating #AI agents
🔗 Link in first 💬⤵️

Repost 🔁 #LLM #Agents
#Meta Starts Using Your Data For Targeted #Ads Oct 7 - no opt out except EU/UK/South Korea. Confirmed: collects interactions with its #AI tools across #WhatsApp #Facebook #Instagram #Threads

Except: religion/beliefs, sexual orientation, politics, health, racial or ethnic origin

More @forbes.com
Meta AI To Start Using Your Data For Targeted Ads—Here’s How And When
Meta has confirmed it will collect interactions with its AI tools to serve targeted ads across Facebook, Instagram and Threads. Here's how, when and who is affected.
www.forbes.com
#MLflow 3.4 adds #MCP server integration — #AI agents can now analyze runs, track tokens & auto-optimize #ML tasks
🔗 Link in first 💬⤵️

Repost 🔁 #LLM #MLOps #LLMOps
Memento — gives your prompts a memory! 🧠
It learns and improves over time, getting closer to the results you’d expect from a fine-tuned #LLM, but without retraining.
Super handy for agents that need to adapt continuously.

Has anyone else played with it yet?
#AI #RAG #PromptEngineering #Agents
GitHub - Agent-on-the-Fly/Memento: Official Code of Memento: Fine-tuning LLM Agents without Fine-tuning LLMs
Official Code of Memento: Fine-tuning LLM Agents without Fine-tuning LLMs - Agent-on-the-Fly/Memento
github.com