Andreas Kollegger
@akollegger
Followers
857
Following
1K
Media
69
Statuses
1K
self referential integrity
Cambridge, England
Joined December 2007
Build better RAG by letting a team of agents extract and connect your reference materials into a knowledge graph. Our new short course, “Agentic Knowledge Graph Construction,” taught by @Neo4j Innovation Lead @akollegger, shows you how. Knowledge graphs are an important way to
58
403
2K
Thrilled to have this course go live! Knowledge graphs expand the context for RAG applications, able to bridge structured and unstructured data to answer a broad variety of user questions.
Learn how to use knowledge graphs to enhance your RAG applications with our new course, built in collaboration with @Neo4j! Explore the basics of knowledge graphs, use Cypher, Neo4j’s query language, build your own graphs, and more. 👉 Join now:
1
1
14
“Any intelligent fool can make things bigger and more complex. It takes a touch of genius - and a lot of courage - to move in the opposite direction.” ~E.F. Schumacher
111
717
3K
“Helping everyone make sense of the world with data,” sounds vaguely familiar
Why do I like @observablehq? This presentation explains it. But also, this presentation was made using Observable, using code written by someone else. I added audio narration, using code I wrote myself. That's why I like @observablehq. https://t.co/dkPpRwJSXV
0
0
2
D3 gave Neo4j its first steps into graph visualization, forever changing conversations from what’s a graph to I want that. #nicetweetsforD3
0
4
12
"Data driven" is annoying when asked to prove people sleep every day, with contrary-evidence of "well, they're never sleeping in our shop."
1
0
0
Can you put https://t.co/aBlnQtl9G5 in a https://t.co/py2kJoRIF9 ?
Quiver: An *extremely* nice looking web app for creating commutative and pasting diagrams, created by @varkora
https://t.co/MAUlaTYVUL
https://t.co/iGP16CI9tb
@q_uiver_app
0
0
2
making computers better part 1 ‣ why all this is important https://t.co/XxwQ4yEpsW
adamwiggins.com
We live in a computer-embroidered reality. Here are six ways to improve computing in service of human prosperity.
8
14
98
Useful as Guidelines, Harmful as Dogma: 1) Just ship the MVP & iterate 2) Never miss the launch date 3) Justify every feature with data 4) Hit OKRs at all costs 5) Tackle low hanging fruit first 6) Always prioritize what users want 7) Feature parity before differentiation
28
165
1K
Today's project: on my A* implementation page I start with an abstract example. I'm adding a diagram showing how the abstract example could come from a concrete example. https://t.co/02dYUatxKi (Sadly, I wrote the svg by hand…)
6
15
124
6/ great products are the creations of high-trust, high-context teams 👯 Prod, Eng, Design are collectively responsible for shipping great products. You can't build products without mutual trust, a ton of shared context and a culture that makes everyone feel welcome
1
4
23
In good software practices, you version code. Use Git. Track changes. Code in master is ground truth. In ML, code alone isn't ground truth. I can run the same SQL query today and tomorrow and get different results. How do you replicate this good software practice for ML? (1/7)
32
201
1K
Does your team have all 3? If not, which are they missing?
ideas.ted.com
The key elements are intellectual diversity, psychological safety and a purpose that’s worth fighting for. Organizational psychologist David Burkus explains what they are and why they’re so i…
0
55
172
