nowfound

AI · December 15, 2022

NL

Natural language Twitter search using Codex

We built a structured search engine for Twitter called Bird SQL, available at https://www.perplexity.ai/sql. Our search interface uses OpenAI Codex to translate natural language to SQL. Our backend then verifies the SQL, executes it, and displays the results on the web app. This makes large structured datasets like a scrape of Twitter easy for anyone to explore. As background, while working on text-to-SQL as a general problem, we came to believe one of its most powerful applications is as a search tool because: - SQL is hard to write by hand and prone to errors - It allows you…

Visit perplexity.ai ↗Alternatives

What it does

In the maker’s words, at launch

We built a structured search engine for Twitter called Bird SQL, available at https://www.perplexity.ai/sql. Our search interface uses OpenAI Codex to translate natural language to SQL. Our backend then verifies the SQL, executes it, and displays the results on the web app. This makes large structured datasets like a scrape of Twitter easy for anyone to explore. As background, while working on text-to-SQL as a general problem, we came to believe one of its most powerful applications is as a search tool because: - SQL is hard to write by hand and prone to errors - It allows you to iterate quickly if you’re exploring a new dataset - A lot of contextual information that you’d normally have to internalize (e.g. your data’s schema) can be automatically generated and offloaded to the language model Using large language models (LLMs) like Codex to write the SQL for you means you don’t have to worry about the nitty gritty language details, but still benefit from the power of a language like SQL. Also, after seeing the results of the query, you can inspect (and if necessary, change) the SQL. The lack of this sort of explainability of the query result is one of the more notorious challenges of returning the output of an LLM directly to the user. Additionally, using LLMs in this way makes these kinds of queries over structured data accessible to people who know little or no SQL. While Bird SQL shares significant infrastructure with our more general LLM-powered search engine over unstructured data (Ask Perplexity - https://perplexity.ai[1]), the two approaches and their respective challenges are quite different. For example, the type of models are different (GPT3.5 vs Codex), obviously the model prompts have different structure, and how to verify model output when its text vs when it's code is different. We are currently exploring ways to combine the two approaches, such as using the results of retrieving information from a structured source (as in Bird SQL) as one of the inputs for the LLM to interpret or summarize (as in Ask Perplexity). We would love to hear your questions, suggestions, and feedback! [1] https://news.ycombinator.com/item?id=33897884

Does the same job

all alternatives →
  • ShapedQLJan 2026 · ▲211

    The SQL engine for search, feeds, and AI agents

  • Text2Query2025 · ▲103

    Turn plain language into powerful database queries

  • Kueri Platform2016 · ▲121

    A complete platform for integrating natural language to SQL

  • Zing2022 · ▲119

    Query and visualize your data using OpenAI from your phone

  • DS
    Describe SQL using natural language, and execute against real data2021 · ▲66

    I played around with GPT-3 to build this demo. Select a public BigQuery dataset and describe your query in natural English, then edit the generated SQL as needed and execute it. https://app.tabbydata.com/sql-assistant-demo

  • Dataherald2024 · ▲126

    Embed NL-to-SQL into your product

More ai this month

the category →
  • I trained a 125M model to autocomplete piano on-device▲598

    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▲585

    Automate your B2B brand going viral, with new media creators

    AI · 18d ago · company-app.joinastute.com

  • Grok Bot▲547

    AI teammates that you can give real work to

    AI · 25d ago · x.ai

  • Needle2: 14MB agentic LLM for phones, wearables, smart home and robots▲537

    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

  • Coldtea.ai▲512

    Make your software self-driving

    AI · 30d ago · coldtea.ai

  • Soloop▲472

    Approval-first Agent OS for solo founders

    AI · 30d ago · soloop.io

Launched alongside, December 2022

the whole month →