GitHub Copilot for SQL
Copilot has proven quite handy as a "smarter Intellisense" tool for me. It often accurately predicts the arguments I need for functions, including their types. Occasionally, when I start typing `a.map(`, it automatically fills in the transformation code I intended to write. The key enabler is the ability to ignore most suggestion. They don't get in my way. The simpler the task I'm working on, the higher the likelihood it gets it right. This is where I get the most benefit from Copilot, as I already understand what I'm coding, why I'm coding it, and how it should appear. Copilot sometimes…
What it does
In the maker’s words, at launch
Copilot has proven quite handy as a "smarter Intellisense" tool for me. It often accurately predicts the arguments I need for functions, including their types. Occasionally, when I start typing `a.map(`, it automatically fills in the transformation code I intended to write. The key enabler is the ability to ignore most suggestion. They don't get in my way. The simpler the task I'm working on, the higher the likelihood it gets it right. This is where I get the most benefit from Copilot, as I already understand what I'm coding, why I'm coding it, and how it should appear. Copilot sometimes saves me the 5-30 seconds it would take to write it manually. Over the course of a day, these saved moments add up. But I always found myself turning to ChatGPT for help writing SQL. I would 1) run a query to get the schema. 2) Feed schema to ChatGPT. 3) Ask GPT to write my query. 4) Paste query into client. Solid results, but the workflow was not great with lots of copy and paste and multiple apps. I work at Hoopdev, an infrastructure access gateway and we happen to have a build-in web client. Therefore our language server has a REST API that we could hook to GPT4. But our first attempt failed. We tried the Ask AI box UX, which only solved half the problem: providing the schema as context. Not getting the way was still missing. The bad suggestions required another prompt to get fixed. Therefore we went on to build the copilot experience [0], but the opposite of Github copilot happened: most suggestions are good. This is because the only context for writing SQL is the database schema and what you want to do, all the model has to do is translate your english query to another language (SQL) using a small context (schema), as opposed to full codebases. Hoopdev's client is web-based, because of that you can try it in 15 seconds. After signing up you get a demo Postgres ready to be queried in the web [1]. If you don't wanna sign up, run it yourself and use your own OpenAI keys [2]. Please let me know what you think. any feedback is welcome. Is this helpful? Do you know/use a better alternative for SQL? 0. https://hoop.dev/blog/introducing-an-ai-powered-query-builde... 1. https://hoop.dev/start 2. https://hoop.dev/docs/deploy/AWS
Does the same job
all alternatives →
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 16d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 26d ago · cactuscompute.com
Launched alongside, June 2024
the whole month →
La Growth Machine▲1,216Create personalized, multi-channel conversations at scale
AI · 2024 · lagrowthmachine.com
PyjamaHR▲1,132Hiring on autopilot. The AI applicant tracking system (ATS).
Work · 2024 · pyjamahr.com