Not sure whether Codex or Devin are saving me time. The workflow is off. They are made to push to the remote, which completely breaks the workflow (working directory -> staging -> remote). So you push or make a pr, pull it down, check whether it's working or ...

Garmin (yesterday) : you did a hard workout, minimum recovery 48 hours Garmin (today) : time to run 24 minutes at 5k threshold pace

OpenCode is really stepping it up. I love the sidebar stuff. Really plays well with nvim.

i dream of a world with a single consistent markdown format

@transistor.fm When I put and episode in "Draft" and then schedule it a few days out, the start date in the analytics should still be the day I post the episode not the day I created the draft

@howaiisbuilt.fm is back. most AI conversations obsess over capabilities. This one focuses on constraints - the right ones that make AI actually useful rather than just impressive demos. open.spotify.com/episode/1MbZ...

Why does only Zed have autocomplete in the chat

Thanks buddy

Thanks to @nicolay.fyi for giving me the opportunity to talk about @trustgraph.bsky.social on How AI Is Built! Labels such as "facts", "observations", and "assertions" take on new meanings when we begin to consider time. Click 👇 to watch the full episode! 🎙️ youtu.be/VpFVAE3L1nk?

Does anyone have a clue about what data structure is used for Google Docs? Suggestions, comments, collaborative editing,...?

EU mobilizes 200 billion Euros for AI. Unless we have a massive political change that money will just go to waste. For innovation to happen, we don't need money first, but deregulation. We cannot work on breakthrough technologies, when we are at constant fear of being sued.

RAG is dead. Long live RAG. LLMs suck at long context. This paper shows what I have seen in most deployments. With longer contexts, performance degrades.

foqos.app "Physically block distracting apps" Grab a NFC tag, lock your app, throw NFC tag off balcony, start coding.

I am dreaming up an Excel backend. Excel as the frontend, Iceberg as the backend. But how do you handle formulas, pivot tables, ...? How do you malformatted sheets? You probably have to put restrictions on what you can do with Excel so it's valid to be stored.

human clipboards is the best description I have heard for people who just copy and paste code and errors from and to ChatGPT

The more I learn about MLOps, the more it's ops first, ML second. Start with: - Build serverless apps - Write infrastructure code - Set up CI/CD - Run VMs & databases - Deploy Kubernetes ML comes after you can build things that work.

TIL: In Postgres, you can dollar quote strings with a lot of ' so you don't have to double each '. $$Dianne's horse$$ $SomeTag$Dianne's horse$SomeTag$

That's probably one of the best websites I have ever seen and one of the best experiences on the web www.stripe.press/poor-charlie...

Dropping some new episodes on @howaiisbuilt.fm . Links below.

New podcast with Alex Garcia on search in sqlite

TIL: Ionos has 1€ email for your custom domain.

Coming back and back again to the same question: How can I make it easy to experiment? With genAI it will be easier to generate alternatives and options. So for almost anything, I want to generate a lot of options, vote for the best one, test it in a sandbox, A/B test, deploy it... (rollback)

Really interesting paper. Machine Learning: The High-Interest Credit Card of Technical Debt static.googleusercontent.com/media/resear...

That's surprisingly on brand.

Good pipes beat perfect models.

What are some of the top applied ai / data engineering blogs (companies)? I would put prefect up there, temporal is great, modal as well.

Best youtube channel, prove me wrong www.youtube.com/shorts/-E9-Y...

Google will win the GenAI but lose the search.

Testing the Bluesky API with curl! 🚀

Developers treat search as a blackbox. Throw everything in a vector database and hope something good comes out. Throw all ranking signals into one big ML model and hope it makes something good out of it. You don’t want to create this witch’s cauldron. New episode on @howaiisbuilt.fm

I spent too much time on logging today. nicolay.fyi/til/setting-...

The biggest lie in RAG is that semantic search is simple. The reality is that it's easy to build, it's easy to get up and running, but it's really hard to get right. And if you don't have a good setup, it's near impossible to debug. One of the reasons it's really hard is chunking.

Data tragedy: ODBC supports *columnar* data retrieval (SQL_BIND_BY_COLUMN) but using it actually *slows things down* for large result sets, from SQL server of all things

I am looking for someone hacking with knowledge graphs. Grateful for any intros / recs.

2024 in review. nicolay.fyi/posts/person...

The search stack we deserve: ✅ Document understanding ✅ Query understanding ✅ Exact matching when needed The search stack we have: "lol just embed it"

All I want for christmas is a cluster of blackwells.

@merve.bsky.social Can you show the Amazon people how to use a VLM to do the handwriting recognition. That's atrocious.

Getting ads for the remarkable pro while watching a review on the new Kindle Scribe. Someone got their targeting down.

Focus on data quality, pick the right tools, and let AI do the heavy lifting.

Monday: 'I'll use design patterns and create good abstractions!' Friday: 'log.debug() is my design pattern'"

Awesome research about recovery oriented computing from @nicolay.fyi . I think it makes a ton of sense and it's also very aligned with a lot of modern development where we invest in fast time to production with good tooling for fast rollbacks/recovery/etc.

Ninety-Ninety Rule: "The first 90 percent of the code accounts for the first 90 percent of the development time. The remaining 10 percent of the code accounts for the other 90 percent of the development time."

My perfect programming language: - Clean and straightforward syntax of Go - Rust compiler & package manager - Erlang's "Let it crash" - Python's ecosystem

Looking for an Azure engineer for a project (freelance). Multi-month contract, workload roughly 10-20h / week. Data pipelines, schema discovery, schema management. DMs are open.

I think it's quite interesting that when you chat with Claude or ChatGPT on "AI", it by default assumes LLMs. "Higher memory usage allows for larger context windows and more complex operations" Really mirrors the general population.

I see so much FUD about the future of sw engineering, mostly from non-devs. Along the lines of “soon anyone can spin off AI agents in bulk that act as hundreds of devs.” A false premise. Just open your airline app that is built by ~hundreds of devs over 10+ years

"Every program attempts to expand until it can read mail"

"Pick your core business competencies and goals, and do those in house. If you’re a software company, writing excellent code is how you’re going to succeed. Go ahead and outsource the company cafeteria and the CD-ROM duplication....

A new workhorse is something I can get down with. Will report on the weekend how it holds up.