SearchGPT, Perplexity’s top rival, saves you massive time

Agentic Progress: GitHub's Spark, Writer's AI Studio, Build own Agent

Author

Martin Musiol
November 01, 2024 • Estimated Reading Time: 14 minutes

In partnership with

The world is becoming more agentic. Learn next

✅ SearchGPT saves you massive time. (10X Google)
✅ Writer for swift app development
✅ Which coding language is most succinct
✅ GitHub’s Spark, a productivity masterpiece for app development
✅ Workshop: I’ll teach f2f - Build your own agent and enjoy working with it

OpenAI released SearchGPT—take advantage of the hot search market

SearchGPT is Perplexity AI’s main rival in search. I haven't used Google in ages—traditional Google Search is fading, though Google still holds ground with Gemini.

Don’t waste time with old research methods—use these advancements by being another 30+% more efficient. I’ll happily keep you posted with this newsletter. 🤎 

The image is a screenshot of a digital interface, likely from a travel or dining recommendation platform. Here is an in-depth description: **Top Section:** - At the top, there's a search query box with the text "where can i grab dinner in positano on friday night?" indicating the user's intent to find dining options in Positano for a Friday night. **Middle Section:** - Below the search query, there's a header text: "Planning a dinner in Positano on a Friday night offers a delightful combination of exquisite cuisine and stunning coastal views. Here are some top recommendations to consider:" - This header provides context about the dining experience in Positano. **Restaurant Listings:** 1. **La Tagliata:** - **Description:** "Nestled in the hills above Positano, La Tagliata provides panoramic views of the Amalfi Coast. This family-run restaurant is renowned for its authentic Italian dishes, particularly handmade pastas and grilled meats. Due to its popularity, especially on weekends, it's advisable to book in advance." - **Button:** There is a button labeled "LA TAGLIATA" for more information or booking. - **Image:** On the right side, there's an image of a dining setup with a scenic view of the mountains and sea, possibly taken from the restaurant itself. The table is set with plates, cutlery, and glasses, and there's a bottle of wine on the table. The setting appears to be outdoors with a pergola-like structure. 2. **Il Ritrovo:** - **Description:** "Located in the heart of Positano, Il Ritrovo offers a cozy ambiance with a menu featuring fresh seafood and traditional Neapolitan cuisine. The restaurant is praised for..." - The description is cut off, so the full text is not visible. - **Image:** There is an image to the right, showing a dining table set for a meal with a view of the sea. The table has plates, cutlery, and glasses, and the setting appears to be on a terrace or balcony with a curtain providing some shade. **Overall Layout:** - The interface uses a clean, white background with black text for readability. - There is a small icon, possibly a logo, to the left of the header text. - The restaurant names are in bold, making them stand out. - The images are placed to the right of the descriptions, providing visual context for each dining option.

SearchGPT has a conversational interface, cites its answers, and has reliably up-to-date information.

The advantage it has over Perplexity AI (that I am very fond of) is that ChatGPT knows you and can tailor your information!

The image is a screenshot of a digital interface, likely from a productivity or habit-tracking application. Here is an in-depth description: **Top Section:** - There is a small icon on the left, which appears to be a circular symbol with some lines or shapes inside it. - Next to this icon, there's a label "Memory updated" with a black bookmark icon, indicating that some information has been saved or updated. **Middle Section:** - A grey rectangular box contains the text: "Martin has started bouldering and aims to improve over the next 12 months." This suggests that the user named Martin has set a goal related to bouldering for the next year. **Bottom Section:** - Below the grey box, there is a button or link labeled "Manage memories" with a right-facing arrow, indicating that clicking it will take the user to a section where they can manage or view their saved memories or goals. - Partially visible at the bottom is another section that seems to be a list or guide, with the visible text: "2. Climb Regularly: Aim for 2-3 sessions a week, with..." The interface appears to be designed to help users track and manage personal goals or habits, with the example here focusing on bouldering.
own animation

Writer RAG tool: build production-ready RAG apps in minutes

RAG in just a few lines of code? We’ve launched a predefined RAG tool on our developer platform, making it easy to bring your data into a Knowledge Graph and interact with it with AI. With a single API call, writer LLMs will intelligently call the RAG tool to chat with your data.

Integrated into Writer’s full-stack platform, it eliminates the need for complex vendor RAG setups, making it quick to build scalable, highly accurate AI workflows just by passing a graph ID of your data as a parameter to your RAG tool.

Programming languages have very varying abilities to communicate logic succinctly

The image is a graph titled "Entropy of Top 10 Programming Languages." It displays the distribution of character counts for code written in ten different programming languages. Each language is represented by a density plot, which shows how the character count in code samples varies for each language. **Axes:** - The x-axis represents the "Character Count" of the code samples. - The y-axis is labeled with the names of programming languages. **Languages and their Distributions:** 1. **Java**: The distribution starts around 0, peaks around 200, and then gradually tapers off. 2. **C**: Similar to Java, it starts around 0, peaks around 200, but the tail is slightly shorter. 3. **C++**: The distribution is similar to C, with a peak around 200 and a tapering tail. 4. **Rust**: The distribution shows a peak around 200, with a noticeable spread towards higher character counts. 5. **C#**: The distribution starts around 0, peaks around 200, and has a similar shape to Java and C++. 6. **Go**: The distribution peaks around 200, but the tail is shorter compared to other languages. 7. **PHP**: The distribution peaks around 200 and has a broader spread compared to some other languages. 8. **Ruby**: The distribution shows a peak around 200 and a relatively broad spread. 9. **JavaScript**: The distribution peaks around 200 and has a noticeable spread towards higher character counts. 10. **Python**: The distribution peaks around 200 and has a broad spread, indicating a wide range of character counts in code samples. **General Observations:** - All languages have a peak around 200 characters, indicating that most code samples are of moderate length. - Some languages like Rust and JavaScript show a broader spread, suggesting that their code samples can vary more in length. - Languages like Java, C, and C++ have more consistent character counts with less variation. The graph provides a comparative analysis of how code complexity or length varies across these top ten programming languages.

If you look at character counts of 10 basic programs in each, Java has 2x higher entropy than Python. Java is thus called more verbose.

See here the character count for the five coding examples: Factorial, Quick Sort, Hello World, and Setting up a Web Server. Python consistently has the least characters it uses.

The image is a table comparing the performance of various programming languages across different tasks. The table has eight columns and ten rows, including the header row. Here is a detailed description: **Columns:** 1. **Language**: Lists the programming language being evaluated. 2. **Factorial**: Time taken (in milliseconds) to calculate the factorial of a number. 3. **Quick Sort**: Time taken (in milliseconds) to perform quick sort on a dataset. 4. **Hello World**: Time taken (in milliseconds) to run a "Hello World" program. 5. **Web Server**: Time taken (in milliseconds) to set up and run a web server. 6. **BFS** (Breadth-First Search): Time taken (in milliseconds) to perform a breadth-first search on a graph. 7. **Mean**: The average time taken across all tasks for each language. 8. **Std Dev** (Standard Deviation): The standard deviation of the times taken across all tasks for each language. **Rows:** 1. **Python**: - Factorial: 92 ms - Quick Sort: 195 ms - Hello World: 21 ms - Web Server: 263 ms - BFS: 294 ms - Mean: 173.0 ms - Std Dev: 102.86 2. **JavaScript**: - Factorial: 114 ms - Quick Sort: 264 ms - Hello World: 28 ms - Web Server: 138 ms - BFS: 357 ms - Mean: 180.2 ms - Std Dev: 116.29 3. **Java**: - Factorial: 131 ms - Quick Sort: 595 ms - Hello World: 118 ms - Web Server: 695 ms - BFS: 439 ms - Mean: 395.6 ms - Std Dev: 236.00 4. **C#**: - Factorial: 143 ms - Quick Sort: 482 ms - Hello World: 120 ms - Web Server: 537 ms - BFS: 433 ms - Mean: 343.0 ms - Std Dev: 176.18 5. **C++**: - Factorial: 112 ms - Quick Sort: 545 ms - Hello World: 98 ms - Web Server: 500 ms - BFS: 515 ms - Mean: 354.0 ms - Std Dev: 191.63 6. **C**: - Factorial: 112 ms - Quick Sort: 541 ms - Hello World: 81 ms - Web Server: 589 ms - BFS: 661 ms - Mean: 396.8 ms - Std Dev: 219.95 7. **PHP**: - Factorial: 120 ms - Quick Sort: 358 ms - Hello World: 32 ms - Web Server: 32 ms - BFS: 336 ms - Mean: 175.6 ms - Std Dev: 137.51 8. **Ruby**: - Factorial: 90 ms - Quick Sort: 246 ms - Hello World: 21 ms - Web Server: 222 ms - BFS: 310 ms - Mean: 177.8 ms - Std Dev: 107.83 9. **Go**: - Factorial: 115 ms - Quick Sort: 433 ms - Hello World: 77 ms - Web Server: 217 ms - BFS: 388 ms - Mean: 246.0 ms - Std Dev: 132.83 10. **Rust**: - Factorial: 103 ms - Quick Sort: 431 ms - Hello World: 46 ms - Web Server: 493 ms - BFS: 427 ms - Mean: 300.0 ms - Std Dev: 174.23 The table provides a comparative analysis of the performance of these programming languages in terms of execution time for various programming tasks.

But it is not only how succinct a programming language is but also how readable. Too little is bad, and too much is bad, too.

I choose Python when I can because I think it is in the sweet spot between readability and concise code.

GitHub Spark - Anyone can create micro apps without needing to write or deploy any code with AI

Can you succinctly and coherently describe your idea? Then, you can build products, apps, and other programs.

Throughout this episode alone, you can see the development of agentic workflows.

GitHub Spark is another stab at it. This video shows how intuitive app development with GitHub Spark can be.

As tools grow, their complexity grows. GitHub Spark has an answer for that: NL-based toolchain.

Start with your basic concept and grow complexity naturally through assisted exploration. Interactive previews, revision variants, automatic history, and model selection.

Read the blogpost here.

⚠️ I am on the waitlist. If you want me to demo how to implement it, let me know in the comments or reply to this email.

[Repost] Learn best practices regarding building products/ code with AI!

The AI Summit Seoul has been at the forefront of tech for years.

Only signal, no noise.

This year, I am happy to share GenerativeAI.net's self-developed framework for using current AI tools most effectively in product development.

Additionally, I'll host a hands-on workshop this year. Participate, learn frameworks, build your agent, and enjoy working with an agent at your hand.

The image is a promotional banner for the "AI Summit Seoul 2024," scheduled for December 10-11, 2024, at the COEX Grand Ballroom in Seoul, South Korea. The banner has a high-tech design with a dark blue background and dynamic light-blue wave patterns that evoke a digital or AI-related aesthetic. At the top, in bold white text, it says "AI SUMMIT SEOUL 2024," with the date and venue information immediately below in smaller white text. Below that, the headline in light blue reads, "How AI informs product development, especially co-development with AI Agents," emphasizing the theme of the presentation. Centered below the headline is a circular headshot of Martin Musiol. He has short hair, a trimmed beard, and wears glasses and a light blue shirt, projecting a professional and approachable demeanor. Underneath his image, his name, "Martin Musiol," is displayed in large, bold, light blue text, followed by his title, "GenAI Leader | GenerativeAI.net," in smaller white text. The design conveys professionalism and modernity, aligning with the conference's focus on cutting-edge AI applications and co-development with AI agents.
own

That’s a wrap! I hope you enjoyed it.

Martin

Our webpage