davide petruzzi

Today, I am releasing BlueQuery Tasks ✅ Your personal prompt library for daily / repetitive research tasks on the web or on your folders. Here are some of the things I have specifically built it for 1) company overviews 2) finding key people in a company + their LinkedIn 3) landscape analysis 4) industry analysis 5) Reddit opinions analysis what do you think?

JuliusAI CASE STUDY: 2M users in 18 months. One of the things I love doing with BlueQueryAI is deep researching fast-growing products and startups to reverse engineer how they did it. Yesterday I analyzed Julius, an AI data analyst. I noticed they have a very simple but effective strategy. Here are various points worth taking inspiration from in my opinion. 1) AI focused on a single function. As their founder says: "startups win because of focus" - specialization always beats generalization when you're small. 2) Exploit a weak function of ChatGPT They identified that ChatGPT sucks at data analysis: it doesn't handle real files, doesn't create decent charts, doesn't facilitate collaboration, doesn't connect to databases. Julius solves exactly these pain points. 3) Early acquisition through ChatGPT They got 10-15k initial users from the ChatGPT plugin store before it shut down. A brilliant hack: use the competitor's platform to acquire users who then migrate to your product. 4) Good old SEO After you've tried ChatGPT data analyzer and left unsatisfied, you go looking for alternatives. Maybe on Google or on ChatGPT itself. And if you search "AI data analysis" Julius is the first result. They dominated the most obvious keywords at the right moment, intercepting users disappointed by ChatGPT. 5) Built to share They understood that data analysis is meant to be shared among colleagues and built sharing features into the product from day 1. Every result becomes a word-of-mouth vehicle. 6) Crystal clear target user: Non-technical analysts Just like with no-code tools, they don't target technical users and data scientists, but business owners, researchers, marketers who want insights without knowing how to code. A huge and underserved market. 7) Speed of onboarding and results You connect your database in 2 clicks and start analyzing by asking questions in English. No setup, sales calls, kickoffs etc... 8) Pure prosumer positioning In my opinion, this pattern is emerging more and more in the AI sector: applications built for the role/task and no longer for the company. There are obviously winning enterprise products but I'm increasingly convinced that the most effective AI adoption starts from the bottom and is for personal use. Hope you can get some good ideas from their strategy to grow your products!

Introducing LocalCrawl: a new AI Agent that scrapes Google Places and prepares sales-ready lists of Local Businesses. # This is how it works: ✅ Search in natural language, e.g., Hotels in Rome, Law Firms in Milan. ✅ Scrapes Places and prepares an initial list, including phone number. ✅ Prepares a business overview for each business found. ✅ Enrich with new columns by asking in simple English / Italian. ✅ Save enrichment column prompts as templates for quick reuse. ✅ Export in CSV / XLSX for sales / analysis or upload into your CRM. In this example, I used it to search Hotels in Rimini, then added a column to investigate if the Hotel offers local experiences to its guests. Everything in 2 minutes. # Why we are building it? 1) Working on google places / google maps is manual, tedious and time consuming 2) Apollo, Lusha and company does not cover local businesses well because they scrape most of their lists from linkedin 3) Clay does it, but it is super expensive, slow and imho their google maps scraper sucks a bit We wanted something simple, cheap and fast that does specifically this job very very well. It works very good for any type of local business, from law firm to doctors, hotels and restaurants. # It's still version 1.0, basic but effective. If you wanna try it, link to the web app is in the first comment. I would love your feedback!

@IchorousWings so nichy that am I the only 1 user

How niche is your niche?

Keep noticing that the best AI products are single-task products.

Just released BlueQueryAI Web Data Extraction Ask in natural language, and it extracts data from tens of webpages into a table in seconds. In this demo, it scrapes hundreds of Reddit comments about Tesla stock and classifies them into positive vs. negative outlook. Massively useful for market research

Releasing version 1.0 of BlueQueryAI today—a new Google-style AI search agent. Search in natural language. Find Papers, Patents, Reddit opinions, Videos, Substacks, and News and info in your own Files. The goal is to build the best AI search agent possible. Long road ahead.

Hi guys, I am finally introducing Bluequery Reddit Researcher I've been working on this for over a month now. I wanted to build the best possible Reddit research agent. It searches for the most relevant reddit threads, then reads ALL the comments and extract opinions, sentiment and insights from social discussions. It will help you research pain points, products and competitors. Hopefully, it will make market research way more fun and faster.

Hi everyone! I've just deployed a useful new capability to BluequeryAI: it can now deep research with focus on scientific papers - Connects to Google Scholar, Arxiv, and Semantic Scholar. - Finds relevant scientific papers based on your prompt. - Analyzes and prepares a quick summary and data overview for each one (authors, findings, methodologies, etc.). - Prepares an in-depth scientific review of the field with major insights. Unlike classic deep researchers, BlueQuery is a full-on text miner. So, it works really well for search & extract tasks like this one. The goal is to provide an overview of the 10–30 most relevant papers on a topic in 5 minutes and significantly speed up research time. I will work on research for patents, reddit opinions, and companies soon

competitive research example -> app.bluequery.ai/share/6bfac1a3…

Today, I am finally introducing "competitive landscape deep research" in BluequeryAI. The most requested specialized version of deep research in this community 💪 1. Deep search for the most relevant companies in a niche. 2. Prepare an in-depth report and overview of the market. 3. Prepare a profile for the top 10 most relevant companies (including funding, leadership, technology, etc.), and a full list of all players in the market (up to 50 players). Let me know what you think! an example in the first link

Hi guys! Here's an example of competitive research you can fully read to understand the type of output Bluequery produces → app.bluequery.ai/share/6bfac1a3…

Today, I am introducing competitive landscape deep research in BluequeryAI. Here is how it works: 1. Deep search for the most relevant companies in a niche. 2. Prepare an in-depth report and overview of the market. 3. Prepare a profile for the top 10 most relevant companies (including funding, leadership, technology, etc.), and a full list of all players in the market (up to 50 players). 4. Find top white papers, charts, and landscape maps related to that market. 5. Prepare summary tables for market segmentation and information recaps. If you find it useful, I’m very excited for you to try it out — let me know what you think! an example in the first link

@IfyGenius that's great idea

@davidepetruzzi_ How about hacker news, great for tech and startup opinions.

hi guys, I am building an AI research assistant for social discussion on online forums (AI thread search + deep research) what niche forum would you like to include? goal is to quickly extract opinions about topics

@N_Nilashan yes, but how do you monetize? you will still be without money at the end of that

I realized marketing does not exist. There is building and there is selling. Marketing is a broad word, that it is hard to truly understand for me. I am a simple human, I build a product / service that solves a need. I contact potential target customers to propose my product / service.

Everybody keeps saying learn marketing. No roadmap. No resource. What's that one marketing advice you got that did it for you? Devs are listening...

@BuiltByTim Notion

Where do you save your favorite prompts?

@aryanrajputvats I think Make and n8n are both very good for AI automations workflows