Ultra-fast, Multi-tenant, and Open-source Graph Database Powering Generative AI.
Tel Aviv, Israel
Joined July 2023
- Tweets633
- Following94
- Followers2.1K
- Likes886
Pinned Tweet
We’re teaming up with @WeMakeDevs for Graph Hacks: Context for AI Agents 🚀
📅 Oct 15–18
🌎 Online
🏆 $20K in prizes
Build AI agents that use graph-powered context.
Register:
wemakedevs.org/hackathons/fa…
FalkorDB retweeted
7,000+ hackers, $20,000 in prizes, and four days of building. Graph Hacks has officially wrapped up! 🎉
Builders came together to create AI agents exploring connected data, persistent memory, and multi-agent coordination, putting FalkorDB to use in projects that tackled different challenges.
Thank you to everyone who participated and made this edition happen, and to @falkordb for collaborating to bring Graph Hacks to life.
Want to take on another challenge? Join No Hands: The Wispr Challenge, where you build a tool to solve a real problem in your work using your voice instead of typing.
Register here: wemakedevs.org/wispr
everyone shipped agent memory this week 👀 @googlecloud @digitalocean too
but memory without connections is just a junk drawer
on GraphRAG-Bench (ICLR 2026), our GraphRAG scored 29% higher accuracy than vector RAG
is your agent’s memory still a junk drawer?
We’re excited to introduce the Muse Wonder series with Muse Wonder and Muse Wonder Full, our most capable models yet. Both are frontier models, built to deliver state-of-the-art performance for personal and professional agentic tasks.
Muse Wonder is available today via the Meta Model API and Muse Code. Muse Wonder Full is available today for Muse Code subscribers. Muse Wonder will also soon power experiences in our personal agent product, Muse, rolling out first to subscribers.
🚨 Last day to sign up for the FalkorDB Hackathon!
Got an idea? Build it with FalkorDB and show us what you can create.
Registration closes today 👇
wemakedevs.org/hackathons/fa…
Missed our webinar on finding the shortest path? The recording is now live. 🎥
See how Dijkstra, A* and CCH compare when graphs get big, and why shortest-path performance gets much harder at scale.
Watch here 👇
falkordb.com/news-updates/sh…
FalkorDB retweeted
If you needed a reason to finally build that AI agent idea, an iPhone 18 Pro should do it. 👀📱
The best project in the Agents That Act on Connected Data track at Graph Hacks wins one. Build an AI agent with @falkordb that understands how data connects and uses those relationships to make decisions and take action.
Graph Hacks runs October 15–18, completely online and open worldwide. You can build solo or with a team of up to 4.
There’s $20,000 worth of prizes across Graph Hacks, with a PS5 with GTA VI, an M6 Mac Mini, AirPods 5, and more also on the prize list.
Register here: wemakedevs.org/hackathons/fa…
FalkorDB retweeted
I'm excited to be hosting #SFTechWeek Agent Memory Night on Oct 8. ⚠️ MAC MINI giveaway, MEMORY included 🧠
Come hear from Vasilije Markovic @tricalt (Founder & CEO @cognee_), Kyle Cabigon @kylecabigon (PM @twelvelabs), Roan Weigert @roangws (DevRel Lead @gmi_cloud) and me (Developer Relations Lead @FalkorDB).
partiful.com/e/rJwjMXqVKMfKT…
@Techweek_
We cohosted Hack for Humanity SF with @GeminiApp and @Photon, part of @AICollectiveCo's global hackathon across 130 communities in 52 countries.
199 builders. One day. Real community problems.
Our new blog breaks down what they built and how 👇
falkordb.com/blog/how-builde…
FalkorDB retweeted
FalkorDB 6.0 is out! 🚀 First release on our new Rust engine 🦀
⚡ 39% fewer CPU instructions
🛠️ 226 issues resolved
Read more here: falkordb.com/blog/announcing…
FalkorDB 6.0 is out! 🚀 First release on our new Rust engine 🦀
⚡ 39% fewer CPU instructions
🛠️ 226 issues resolved
Read more here: falkordb.com/blog/announcing…
1. Dependency graphs: node A depends on node B means an edge from B to A.
2. C4 model drills top-down: Context, Containers, Components, Code.
3. Good visualization cuts onboarding time by making architecture legible, not just documented.
falkordb.com/blog/code-visua…
1. Vector search finds similar text, not related facts.
2. Graph-only retrieval can't rank by relevance without embeddings.
3. FalkorDB stores embeddings on graph nodes: one Cypher query, traversal + vector search.
falkordb.com/blog/what-is-hy…
1. VRP dates to 1959 — Dantzig & Ramser's truck dispatching problem.
2. Capacity, time windows, backhauls: same map, three different solver problems.
3. Stochastic VRP needs 15-25% more fleet distance to hold a 95% service level.
falkordb.com/blog/vehicle-ro…
1. RedisGraph deprecated Feb 2025 — support ended, repo frozen.
2. FalkorDB is the direct successor: same Cypher, same Redis persistence.
3. Migrate via RDB: SAVE from RedisGraph, mount into FalkorDB. No rewrite.
falkordb.com/blog/redisgraph…
1. Text-to-Cypher turns "Who are Eve's friends?" into a real query — no MATCH/WHERE.
2. One Docker cmd: FalkorDB + browser + API + MCP.
3. MCP plugs into VSCode/Claude Desktop for graph queries in your editor.
falkordb.com/blog/text-to-cy…
1. Vector RAG retrieves similar text, not connected facts — why multi-hop queries collapse.
2. Graph vs vector-only, 4 domains: correctness ~50%->80%+.
3. Route by entity count: 1 -> vector, 3+ relational -> graph.
falkordb.com/blog/ai-agent-m…
