5 Powerful AI Tools for Developers

5 Powerful AI Tools for Developers
Photo by Douglas Lopes / Unsplash

Generative AI models like ChatGTP have become commonplace in our lives and work. They're also helpful for developers to streamline their work, enhance coding skills, and implement powerful AI features inside apps.

Here are five powerful AI developer tools I recently discovered.

Perplexity.ai

Perplexity is a powerful AI search engine. It searches the web and provides a conversational, verifiable, sourced answer. Those who have used Arc search will find Perflexity a familiar interface. Ask it any question, and it offers real-time and accurate information.

In addition to source transparency and a conversational interface similar to ChatGPT, Perplexity also has contextual memory that remembers the context of your previous queries.

The Perplexity API provides applications with powerful search capabilities, meaning they can respond to natural language queries with up-to-date information. It is compatible with the OpenAI library by changing the base_url.

from openai import OpenAI

client = OpenAI(api_key=YOUR_API_KEY, base_url="https://api.perplexity.ai")

response = client.chat.completions.create(
    model="llama-3-sonar-large-32k-online",
    messages=[
        {"role": "system", "content": "You are an AI assistant."},
        {"role": "user", "content": "How many stars are in the universe?"}
    ]
)

print(response)

Using the Perplexity API with Pyhon

ScrapeGraph

ScrapeGraph streamlines data extraction to build AI capability in an application. The good 'ol days of scraping meant loading pages and querying xpath to extract data. Today's LLMs can perform scraping tasks much more efficiently with a single prompt.

With ScrapeGraph, provide any website and ask it a question like extract all social media links. It will respond with an object of links to social media profiles found on the page.

smart_scraper_graph = SmartScraperGraph(
    prompt="Find some information about what does the company do, the name and a contact email.",
    source="https://scrapegraphai.com/",
    config=graph_config
)

ScrapeGraph also comes bundles with proxy rotation and rate limiting features designed to maximize data extraction without getting blocked.

LibraChat

For those working with custom models or wanting to experiment with and compare different LLMs, there's LibraChat. It's an open-source ChatGTP clone that serves as a hub for all of one's AI conversations.

Users can connect any LLM to LibraChat and switch at will. For example, it's possible to use gtp-4-version-preview to test the most cutting-edge GTP model and switch to c4ai-command-r-v01 reasoning, summarizing, and answering questions.

In simple terms, LibraChat is your browser for AI models.

Libra chat interface

Open Router

If LibraChat is the browser, OpenRouter is the marketplace. OpenRouter aggregates over 250 models, making them accessible through a single interface or API.

Rather than signing up and paying for different models, OpenRouter includes hundreds of models in a single, centralized hub. For developers, a single API key provides access to dozens of model providers for versatile integration of AI features in any app.

Cursor

Cursor gives us a glimpse into the future of programming. A fork of VS Code, Cursor is an IDE that enhances developers' coding.

The ultra-powerful "Chat with your code" feature allows you to make inline queries on your code for insanely fast prototyping and development. In seconds, a simple query like rewrite this code from javascript to python generates code that would have usually taken hours to write and test.

Cursor even makes predictions and will show the code it thinks should come next. Just hit Tab and it implements the changes.