Chris Bunk

CPTO @ FastSpring - Agentic Engineering Enthusiast
My Tweets are my own

[FastSpring @FastSpring]

More control over revenue. More control over player data. More control over the path from discovery to purchase. More control over the tools they rely on.

We unpack all five in our latest blog from our Head of Gaming, Chip Thurston.

[FastSpring @FastSpring]

India has always had a subscription culture.
What changed is the payment layer.

UPI has become the dominant force moving those offline, cash-based habits into digital recurring payments. https://t.co/jY3KJhLGcU

[FastSpring @FastSpring]

Over 90% of web store purchases come from players who made a D2C purchase in the last 30 days.

That stat from our Head of Gaming, Chip Thurston, tells you everything about where the D2C opportunity lives: not in acquisition, but in repeat behavior. https://t.co/yvBfd6HvS4

[FastSpring @FastSpring]

We're thrilled to welcome LLR Partners and to continue building alongside Accel-KKR as we accelerate product innovation, expand our go-to-market, and help more technology companies offload the complexity of selling globally. https://t.co/Yil0Jus3Sz

[FastSpring @FastSpring]

Stop renting user relationships from Google Play.

We released our definitive guide on integrating app2web monetization into your Android app's user acquisition strategy.

Download the guide → https://t.co/hvrjfWkDo1 #AndroidMakers26

[Chris Bunk @chrisbunk]

Thanks to Bobby Tahir for having me on his podcast to talk about my GenAI Sabatical and three buckets of GenAI value creation - https://t.co/andnah90yV

[Chris Bunk @chrisbunk]

@mstockton Agree with all of this. You state it very well. Check out this podcast I recently was on, it echos a lot of what you have been saying and details my past year taking an AI sabatical.
https://t.co/andnah90yV

[Chris Bunk @chrisbunk]

The Golden Rule for Learning Agentic Development with Others:
Any question you would ask of somebody in the group ask about what your trying to do, try asking a LLM first.

[Chris Bunk @chrisbunk]

"Won Nobel Prize" and "founded company" aren't semantically similar.
But they're one hop away from the same person node.
This is why vector search alone fails—and why Graph RAG changes everything for AI-assisted development. https://t.co/YDLGO22zgt

[Avi Chawla @_avichawla]

RAG vs. Graph RAG, explained visually!

RAG has many issues.

For instance, imagine you want to summarize a biography, and each chapter of the document covers a specific accomplishment of a person (P).

This is difficult with naive RAG since it only retrieves the top-k relevant chunks, but this task needs the full context.

Graph RAG solves this.

The following visual depicts how it differs from naive RAG.

The core idea is to:
- Create a graph (entities & relationships) from documents.
- Traverse the graph during retrieval to fetch context.
- Pass the context to the LLM to get a response.

Let's see how Graph RAG solves the above problem.

First, a system (typically an LLM) will create a graph from documents.

This graph will have a subgraph for the person (P) where each accomplishment is one hop away from the entity node of P.

During summarization, the system can do a graph traversal to fetch all the relevant context related to P's accomplishments.

The entire context will help the LLM produce a complete answer, while naive RAG won't.

Graph RAG systems are also better than naive RAG systems because LLMs are inherently adept at reasoning with structured data.

👉 Over to you: Have you used Graph RAG in production?

[Chris Bunk @chrisbunk]

Agree with a lot of this
https://t.co/fbAkkdxomn

[Chris Bunk @chrisbunk]

@jeanqasaur SRE Playbooks

[Chris Bunk @chrisbunk]

@bdickins At the same butcher, at the beginning of lockdown Covid, I got waygu filets for cheaper then chuck steak at stop and shop. It’s all about the supply and demand. That said I would expect Ribeyes, especially in the summer to be constantly in demand. I do love Skirt Steak though

[Carl Vellotti 🥞 @carlvellotti]

I freaking love this PM visualization.

The closer you look, the more awesome it gets.

[Mark Tenenholtz @marktenenholtz]

I am at least 3-5x more productive using ChatGPT to code.

Not only am I faster writing code I'm familiar with, but I've even shipped apps in tech stacks I'd never used before.

Here's my process, the prompts I use, and why it all works:

[Chris Bunk @chrisbunk]

@CubicleApril Works every time:
https://t.co/JuQwscsWK6

[Chris Bunk @chrisbunk]

#ChatGPT Use Case: I took my 9/7/3yo Kids and asked them for prompts for a story then read the output to them. Better stories (personalization helped) them most stories plus can do things like "tell in style of Dr Seuss"

[Dave Kline @dklineii]

Your boss is a coward and it's killing your career.

Here's the honest feedback you need to hear:

[Engine by MoneyLion @enginebyml]

We are honored to be named in Built In's 2023 Best Places to Work list. Interesting in working with Even? Please check us out here: https://t.co/H9HDigwMF3

[Chris Bunk @chrisbunk]

LLM Idea: Make a news site that scans in new articles for a range of sources and reproduces the content focusing on facts and without click bait headlines, editorial or political bias.

[Santiago @svpino]

11 ways ChatGPT saves me hours of work every day, and why you'll never outcompete those who use AI effectively.

A list for those who write code:

1 of 16