Seems like we are hearing a lot about GraphRAG these days, but there are lots of questions: what is it, is it hype, what is practical? One of our all time favorite podcast friends, Prashanth Rao, joins us to dig into this topic beyond the hype. Prashanth gives us a bit of background and practical use cases for GraphRAG and graph data.

This week Daniel & Chris hang with repeat guest and good friend Demetrios Brinkmann of the MLOps Community. Together they review, debate, and poke fun at the 2024 Gartner Hype Cycle chart for Artificial Intelligence. You are invited to join them in this light-hearted fun conversation about the state of hype in artificial intelligence.

There’s so much talk (and hype) these days about vector databases. We thought it would be timely and practical to have someone on the show that has been hands on with the various options and actually tried to build applications leveraging vector search. Prashanth Rao is a real practitioner that has spent and huge amount of time exploring the expanding set of vector database offerings. After introducing vector database and giving us a mental model of how they fit in with other datastores, Prashanth digs into the trade offs as related to indices, hosting options, embedding vs. query optimization, and more.

Ernesto Falcon, writing for the EFF:

[wireless carriers] are only trying to focus our attention on 5G to try to distract us from their willful failure to invest in a proven ultrafast option for many Americans: fiber to the home, or FTTH.

He goes on to break down why 5G won’t solve many of our (USA) problems and why it’s better to ignore the hype and ask why we’re falling behind other areas of the world.

There is crazy hype and a lot of confusion related to DeepSeek’s latest model DeepSeek R1. The products provided by DeepSeek (their version of a ChatGPT-like app) has exploded in popularity. However, ties to China have raised privacy and geopolitical concerns. In this episode, Chris and Daniel cut through the hype to talk about the model, privacy implications, running DeepSeek models securely, and what this signals for open models in 2025.

We’re making some big Changelog changes in 2025, the previously featured Stanford study on ghost engineers doesn’t live up to the hype, Git ingest is a simple service that turns any GitHub repository into a simple text ingest of its codebase, Simon Willison dishes out some hard-earned wisdom he acquired by working at Lanyrd / Eventbrite & Matheus Lima warns us about six mistakes that new managers make.

We are on the other side of “big data” hype, but what is the future of analytics and how does AI fit in? Till and Adithya from MotherDuck join us to discuss why DuckDB is taking the analytics and AI world by storm. We dive into what makes DuckDB, a free, in-process SQL OLAP database management system, unique including its ability to execute lighting fast analytics queries against a variety of data sources, even on your laptop! Along the way we dig into the intersections with AI, such as text-to-sql, vector search, and AI-driven SQL query correction.

Workflow orchestration has always been a pain for data scientists, but this is exacerbated in these AI hype days by agentic workflows executing arbitrary (not pre-defined) workflows with a variety of failure modes. Adam from Prefect joins us to talk through their open source Python library for orchestration and visibility into python-based pipelines. Along the way, he introduces us to things like Marvin, their AI engineering framework, and ControlFlow, their agent workflow system.

In this follow-up to episode #306, “How soon until AI takes my job?”, the gang of (grumpy?) veteran software engineers candidly chat about how their day to day is changing in the midst of improving AI tooling & hype.

Scott Chacon writes up his insider take on GitHub’s success, Sentry wants other companies to take the Open Source Pledge, Benj Edwards used AI to reproduce his late father’s handwriting, Dave Kiss explains the current hype that PHP is getting & Taylor Otwell raises $57 million series A from Accel.

In the midst of the demos & discussion about OpenAI’s GPT-4o voice assistant, Kyutai swooped in to release the first real-time AI voice assistant model and a pretty slick demo (Moshi). Chris & Daniel discuss what this more open approach to a voice assistant might catalyze. They also discuss recent changes to Gartner’s ranking of GenAI on their hype cycle.

Apple announces its “new” style of AI, piku gives you “git push” deployment on your own servers, Dabo Chen rebuilds nanoGPT in a spreadsheet, Mark Seemann thinks you’ll regret using natural keys in your database design & Glyph Lefkowitz describes his grand unified theory of the AI hype cycle.

There’s a lot of hype about AI agents right now, but developing robust agents isn’t yet a reality in general. Imbue is leading the way towards more robust agents by taking a full-stack approach; from hardware innovations through to user interface. In this episode, Josh, Imbue’s CTO, tell us more about their approach and some of what they have learned along the way.

This week on The Changelog we’re talking with Joe Reis about data engineering and the beginning of generative AI. We discuss phone hacking via frequency, the role of a data engineer, this AI hype cycle we’re in, build vs buy, the disconnect between data analysts and the business, ethical considerations around AI-generated content, and more. We also discuss the tension between AI and traditional engineering, as well as the inevitability of AI integration into pretty much everything.

Amal, KBall & Chris convene a “semi-emergency” pod to discuss the recent (deserved) hype over Bun and what it all means for Node’s community, maintainers & users.

They’re joined by Node Technical Steering Committee members Matteo Collina & James Snell who are here to dispel Bun antagonism rumors, discuss the pros & cons of each runtime, explain how Node continues to thrive & even announce a VERY big upcoming feature!

With all the LLM hype, it’s worth remembering that enterprise stakeholders want answers to “why” questions. Enter causal inference. Paul Hünermund has been doing research and writing on this topic for some time and joins us to introduce the topic. He also shares some relevant trends and some tips for getting started with methods including double machine learning, experimentation, difference-in-difference, and more.

What’s the current reality and practical implications of using 3D environments for simulation and synthetic data creation? In this episode, we cut right through the hype of the Metaverse, Multiverse, Omniverse, and all the “verses” to understand how 3D assets and tooling are actually helping AI developers develop industrial robots, autonomous vehicles, and more. Beau Perschall is at the center of these innovations in his work with NVIDIA, and there is no one better to help us explore the topic!

Eric Gregory asks (and answers) himself a question on The New Stack:

For years now, blogs, webinars and white papers have opined that GitOps is the Next Big Thing, yet here a respected voice in the field is saying to tread carefully. So what gives? Do we need to pump the brakes? Is GitOps just a lot of unwarranted hype? Or is there a missing piece of the puzzle here? As in so many things, the answer is: It’s complicated. GitOps can be transformative for some teams, but it’s not a one-size-fits-all solution.

Nader Dabit shares his motivation and experience on recently transitioning to focus on technologies and communities that support the decentralized internet. In this hot topics discussion, we cover all the buzz words you’ve likely heard over the past year. We have honest and nuanced conversations about the world of Ethereum, Cryptocurrencies, NFTs, DAOs, and Web3. Hype or hit? You’ll have to tune in to find out.

Chris Wellons:

Machine learning is a trendy topic, so naturally it’s often used for inappropriate purposes where a simpler, more efficient, and more reliable solution suffices. The other day I saw an illustrative and fun example of this: Neural Network Cars and Genetic Algorithms. The video demonstrates 2D cars driven by a neural network with weights determined by a generic algorithm. However, the entire scheme can be replaced by a first-degree polynomial without any loss in capability. The machine learning part is overkill.

Yet another example of a meta-trend in software: You might not need $X (where $X is a popular tool or technique that is on the upward side of the hype cycle).

A solid interview with Steve Klabnik from the Rust core team:

We sat down with Steve to hear from him first-hand about his professional activities at the moment, the design success of Rust, a little about the “full-stack” development hype, and overcoming burnouts.

Alex Hudson:

Increasingly popular in the last couple of years, I think 2020 is going to be the year of “no code”: the movement that say you can write business logic and even entire applications without having the training of a software developer. I empathise with people doing this, and I think some of the “no code” tools are great. But I also thing it’s wrong at heart.

We had a great dialog about this topic on our 2020 predictions episode of JS Party. I tend to agree with Alex’s analysis, which is in-depth and well-reasoned. What do you think about the “no code” movement? Hype or reality? Somewhere in between?

WIRED’s business unit interviewed Jerome Pesenti, VP of artificial intelligence at Facebook. The major takeaway:

[he] is encouraged by progress in artificial intelligence, but sees the limits of the current approach to deep learning.

Could this be the beginning of the end for this particular AI hype cycle?

GANs are at the center of AI hype. However, they are also starting to be extremely practical and be used to develop solutions to real problems. Jakub Langr and Vladimir Bok join us for a deep dive into GANs and their application. We discuss the basics of GANs, their various flavors, and open research problems.

There’s a lot of hype about knowledge graphs and AI-methods for building or using them, but what exactly is a knowledge graph? How is it different from a database or other data store? How can I build my own knowledge graph? James Fletcher from Grakn Labs helps us understand knowledge graphs in general and some practical steps towards creating your own. He also discusses graph neural networks and the future of graph-augmented methods.

Most people are modest about their contributions in the workplace. We also forget how important our contributions are. Then, when it comes time for recognition, you’ve forgotten, others didn’t notice because they don’t understand all the details and moving parts, and work just moves on. What do you do if/when your work goes unnoticed? Here’s what Julia Evans suggests…

Instead of trying to remember everything you did with your brain, maintain a “brag document” that lists everything so you can refer to it when you get to performance review season! This is a pretty common tactic – when I started doing this I mentioned it to more experienced people and they were like “oh yeah, I’ve been doing that for a long time, it really helps”.

Where I work we call this a “brag document” but I’ve heard other names for the same concept like “hype document” or “list of stuff I did” :).

BONUS — Julia included a basic template for a brag document at the end of the post.

This isn’t just for business executives. It’s good knowledge to have for anyone who has heard the hype around K8S but never any of the potential problems:

This post will cover some hard truths of Kubernetes and what it means for your organization and business. You might have heard the term “Kubernetes” and you might have been led to believe that this will solve all the infrastructure pain for your organization. There is some truth to that, which will not be the focus of this post. To get to the state of enlightenment with Kubernetes, you need to first go through some hard challenges. Let’s dive in to some of these hard truths.

Despite all the recent hype, setting up a new TypeScript (x React) library can be tough. Between Rollup, Jest, tsconfig, Yarn resolutions, TSLint, and getting VSCode to play nicely….there is just a whole lot of stuff to do (and things to screw up). TSDX is a zero-config CLI that helps you develop, test, and publish modern TypeScript packages with ease–so you can focus on your awesome new library and not waste another afternoon on the configuration.

It feels like we’re starting to pass the peak of the hype cycle of microservices. It’s no longer multiple times a week we now see a blog post of “How I migrated my monolith to 150 services”. Now I often hear a bit more of the counter: “I don’t hate my monolith, I just care that things stay performant”

What follows is an excellent rundown of all the advantages that a monolith has over microservices. For a real-world case study, listen to the details of Segment’s transition back to a monorepo.

There’s a lot of wisdom in this post alongside some opinions that I find myself nodding along with:

My strategy for dealing with the absurd pace of change in web development has been as follows: ignore 99% of it and see if it goes away.

While we cover (and talk about) new technologies on a daily basis here at Changelog, that doesn’t mean we adopt everything that hits our radar.

Given the hype cycle, it works pretty well. Mongo isn’t exciting anymore, Angular 1 is dead, CoffeeScript is obsolete, I haven’t heard a word about Meteor since it launched…

These are all specific technologies. But what about the Single Page App pattern in general?

Back in the early days of SPAs, some people argued that it would be faster to only pass the data you need as JSON than to render whole pages. Nearly a decade later, this is almost never true.

He goes on to explain how he’s building a side project SPA-style and all the repercussions of that decision. Really insightful stuff here, please do click through and read for yourself.