The Startup Ideas Podcast - Is Grok 4 Worth $300? I Tested 9 AI Agents to Find Out

Episode Date: July 15, 2025

On this episode, I test Grok 4's capabilities across multiple use cases to determine if it outperforms competitors like OpenAI and Perplexity. I evaluate 9 different agent types by running specific prompts and analyzing the results. Grok 4 demonstrates particular strength in tasks leveraging X data, with standout performance in market research, customer feedback analysis, and negotiation assistance, while showing limitations in areas like visual design. Timestamps: 00:00 - Intro 01:00 - Market research Agent 04:05 - Coding Agent 05:57 - Productivity Agent 10:10 - Fundraising Agent 14:48 - Content MarketerAgent 23:23 - Customer Feedback Agent 27:04 - Negotiator Agent 29:50 - Product Innovator Agent 32:47 - Design Agent 35:17 - Grok Companions Feature 37:19 - Voice mode Demo 38:26 - Final review and conclusions Key Points: • I test Grok 4 across 9 different use cases including market research, coding, productivity optimization, and design • Grok 4's integration with X (Twitter) data provides unique advantages for certain tasks • Performance varied across different agent types, with market research and customer feedback analysis being particularly strong The #1 tool to find startup ideas/trends - https://www.ideabrowser.com LCA helps Fortune 500s and fast-growing startups build their future - from Warner Music to Fortnite to Dropbox. We turn 'what if' into reality with AI, apps, and next-gen products https://latecheckout.agency/ Boringmarketing - Vibe Marketing for Companies: http://boringmarketing.com/ The Vibe Marketer - Join the Community and Learn: http://thevibemarketer.com/ Startup Empire - a membership for builders who want to build cash-flowing businesses https://www.skool.com/startupempire/about FIND ME ON SOCIAL X/Twitter: https://twitter.com/gregisenberg Instagram: https://instagram.com/gregisenberg/ LinkedIn: https://www.linkedin.com/in/gisenberg/

Everyone is talking about GROC4. People are saying it's just outperforming, you know, open AI, outperforming perplexity, outperforming clode. I wanted to test it. So today we're going to test GROC4's design agents, coding agents, market research agents, productivity agents, copywriting agents, customer feedback agents and more. We're going to go through nine prompts. I'm going to give away the prompts.

This is going to help get your creative juices flowing. And by the end of the episode, we're going to see is GROC4? worth playing with. Should you make this a core foundational part of your stack? These are the things we're going to go through and we're going to see if it's worth it. And by the way, these are like, these are the first time I'm actually doing this as well. So we're going to see pretty much live. Well, we're going to see live if this is if this is worth doing. So the first prompt we're going to do is act as a market research agent, analyze competitors,

and we have to pick, we'll just say, productivity apps. Using real-time web and X search, provide a table of top three players, their pricing, user pain points from recent reviews, and untapped opportunities for my startup to differentiate and capture 10% market share. So we're going to go ahead and click that.

is going to be helpful for anyone who has a startup or wants to build a startup. Understanding, obviously, your competitors is going to help you, you know, basically come up with product ideas, how to think about things in a more clear way. So if this works, this would be huge. You know, I think what obviously GROC has that other people don't have is they have X data, right? So the reason why GROC 3 was, you know, when we played with GROC 3, it was like pretty darn good. So I'm not surprised that people are saying that GROC 4 is really, really good because of the unique data that they have. So, you know, from like a UI perspective, you know, it's going, it's browsing. You can see it's searching on X pain points about the TaduS app. I love that you can just

kind of watch what's happening. This has obviously been a trend for a lot of the LLMs, but I love that GROC has this. Look, many ex posts about TaduS don't mention it, TiduS complaints. I mean, it's kind of crazy that you can actually just figure out what are the pain points in a particular industry, and it's going to search the millions of tweets and websites. So now it's creating the table. You can see it puts no shire. first. Wow. It talks about notion slowness. I've actually tweeted about notion slowness. The CEO of Notion actually reached out to me from it because it went viral. And that's why, okay, I'm just like, that's amazing that it has it in there. Untapped afternoon.

Focus on a streamline, lightweight interface to prejudice, speed, and simplicity. Yeah, I agree. Like, I think that like a lightweight version of notion with, you know, a lot less bloat would actually do really well. You know, and then it goes through To-DoS, lack of visual clarity leading to overwhelm, and then what you can do here, the untapped opportunity, introduce advanced visual planning tools like Intuitive, drag and drop. All right, so I'm not going to go through all of this, but the point, you know, this seems like really dialed.

It took a minute and 20 seconds, so not super long. And it did what we said. So so far, so good for prime. Prompt 1, using Grok as a market research agent, it's definitely something that I'm going to be doing more of. Okay, can you use Grok to code your MVP? So we're going to go ahead and copy this prompt here. Using Grok for code mode, so I guess it has a code mode, generate Python code for a simple founder tool like a lead gen bot. So that's our example that we're going to use.

include error handling, integrate with API, EG, Striper Payments, and explain how to deploy it, fix any bugs in the pasted code snippet. We don't have any code snippets yet, but it's good that we can use that prompt if there's any bugs. We'll go ahead and put that prompt in and just tell GROC4 to fix it. So it's going ahead and it's planning the project. So this is similar to how, you know, if you're going to use something like a bolt or a lovable replet,

it's going to do something similar. Now you have this with GROC 4 as another option for you to go and go and do this. So, I mean, here we got the code. I'm not going to go and figure out if we can, you know, deploy, you know, compile and deploy this. But it's pretty cool that it can do that. How fast was that? That took like, I don't know, 20 seconds.

It explains the code. It tells you how to deploy. it. So this is pretty legit. This is pretty legit. And yeah, it took 25 seconds. So, you know, the jury's out, you know, the jury's out around is, like, I haven't deployed this.

So I don't know if it works and stuff like that. But just like quickly looking at it, like, it looks, looks clean. and we're going to go ahead and, you know, go to the next one. So the next prompt is around productivity workflow optimization. So this is just about, you know, increasing your productivity. Let's go ahead and post it in. So as a founder agent, optimize my daily routine. So it says, analyze the schedule.

So I've got to pace my schedule. So let's just say 7 a.m. wake up, you know, 8 a.m. to 10 a.m. deep work. 11 a.m. to 1230, daily standup. 1 p.m. to 3 p.m. to 3 p.m. to 4.30 p.m. meet with my partners. Sam Altman, the co-founder of OpenAI, just said that it is the era of the idea guy. and he is not wrong. I think that right now is an incredible time to be building a startup. And if you listen to this podcast,

chances are you think so too. Now, I think that you can look at trends to basically figure out what are the startup ideas you should be building. So that's exactly why I built ideabrowser.com. Every single day, you're going to get a free startup idea in your inbox and it's all backed by high-quality data trends. How we do it,

people always ask. We use AI agents to go and search, what are people looking for? And what are they screaming for in terms of products that you should be building? And then we hand it on a, you know, silver platter for you to go check out. We do have a few paid plans that, you know, take it to the next level, give you more ideas, give you more AI agents and more almost like a chat TVT for ideas with it. But you can start for free, ideabrowser.com. And if you're listening to this, I highly recommend it. And then it says, suggest AI automated tasks, time saving hacks, and tools to boost output by 2X, pull real-time productivity trends for X. So this is what I'm noticing is a hack for if you want to get the most out of GROC 4. Ask it to go and get data from X. So this is like the biggest takeaway so far, right? Like, you know, if you want the most out of GROC 4, well, it's strength. What is it strength? The strength is that it has a that data from X. So go and, you know, pull that, pull it from it. You know, will GROC 4 be the best coding agent? I don't know because, you know, it's not reliant that much on X data, but something like

this, I bet you it's going to be super, super dialed. So it says your routine as a founder is solid. It's focused on deep work and essential meeting, but it has gaps leading to burnout. And it's probably looking at all those tweets by productivity experts, right? And, yeah, of course, like Huberman and stuff like that. Like wake up, hydrate, you know, light exercise. I mean, this is cool. Oh, my God. This is cool.

Leverage AI to offload 20 to 50% of manual work per X trends where agents handle workflows like research or support. Focus on 80-20 tasks. Use a script to summarize inboxes. Here's a sample Python code for email fetch. This could run daily at 7 a.m. condensing 50 emails into 5 bullets, saving 30 to 60 minutes. This is really cool.

Tools to boost output by 2X. Interesting that it actually gives competitors cursor and replet, which is kind of hilarious. At least that means that it's not super biased. although it does put grok for, you know, hey, use grok, of course. Okay, but this is interesting. Like basically, you know, thinking about how do you use grok to, you know,

look at your calendar and suggest things to do. Obviously, you can go deeper here, ask it to build you scripts, and just an interesting way to, you know, save you time. Okay. So the next prompt is going to be about creating a pitch deck. So it says, I'll just copy it over here. Go back to Grock. So refine this pitch deck script.

I've prepared one as a VC agent. Use reasoning mode to spot weaknesses, suggest data-backed improvements from recent startup funding trends, search X and web. Again, it's really important that we have that there. And predict investor objections with counterarguments. So let's go ahead and I'm going to copy as you can see here, I just pulled this company called Anam. They've, you know, they raised $9 million

seed round. And Anam is a human face for your product. It's this next interface for tech. So it's basically like that movie, Her, but better. Anyways, I grabbed their script. I'm going to go put it on to Grock and just see what it says. And we're going to see what it says, right? Like, you know, if you're going to do a pitch for a VC, obviously you want the best possible pitch. That's why people say, you know, people say, like, try it on your mom first or try it on your grandmother first, like pitch them first. So you get the kinks out. But, you know, if you can get the kinks out earlier and you don't have to pitch your grandmother or mom, that's a win, right?

So the goal here is to get the best possible most dialed pitch based on data. So the data here hopefully is better. So 36 seconds. I'm actually surprised with how fast GROC 4 is. It just feels faster than GROC 3. So as a VC agent evaluating this pitch, I'm going to outline my reasoning on weaknesses, data-backed improvements. Okay. So it says the section like team market and traction are high level without proof. That's huge. I agree with that. There's overly broad use cases. The raise framing feels unmatched. And then it gives you some data backed improvements. And this is, again, this is why we're using GROC for. We're using it because it has the data. Add the TAM and trends. Autonomous agents is the top seed cap.

category. Like, this is really good. And then it gives you the objections and counterargument. So the objection, the space is saturated. Synthesia just raised 180 million at 2.1 billion. How do you compete with that? Well, the counter argument is we're differentiated by pixel by pixel real time rendering and expressive control, enabling low latency interactions that synthesia like video tools can't match. These are really good. And then it gives you a refined script, a refined script around what to do. And then you can go ahead and take this script and post it into a tool like Gen Spark, which can create your AI slides for you, into a tool like gamma, or just do it yourself on Google slides or something like that. I actually wonder, like, can Grok4 create slides? I really like,

the refined script that you created, but don't know if you know how to design slides. I wouldn't be surprised if Grogh 4 could do this just because Gen Spark, Manis, and some of the others can. But let's see. I will say, by the way, that this refined script for this deck is probably, it might be the best I've ever seen.

So this is, you know, in terms of the LLMs, this is really good and it's so quick. Okay, so it could, let's see what it's saying here. I can actually generate or export actual slide files, but it just tells me how they would design it in text. For me, that's not super helpful. So if, but I will say like if you're, this isn't just about raising money, right? Like this is about if you need a deck for sales.

If you need a deck to share with your team, GROC 4 seems to be really, really dialed. All right. We're going to do the content marketing strategy generator prompt. So let's go ahead and put that into SuperGroch. Act as a multi-agent content marketing team. brainstorm a 30-day viral content plan for my niche. We're just going to say ideabrowser.com because it's a startup I'm working on to drive 10,000 new leads. Pull real-time X trends for hot topics, generate 10 post ideas with hooks, SEO keywords, and predicted engagement metrics.

Use thinking mode to evaluate and rank them by potential ROI. And some people, by the way, they might want to know like how to, how did I come up with these prompts? I actually asked GROC to optimize my prompts, and that's why that's why they're so clean and concise and good. And my hope, you know, I always do that when I'm prompting any LLM because I wanted to give it in a way that's going to really work for it. I want to get the most out of the LM, so I usually ask it if it could optimize it. So let's see if, you know, if GROC 4 is good at this, I mean, you would think it would be good at it because, again, X, right? It has all the content or a lot of content.

So you think that GROC 4 would be pretty good at figuring out, you know, what's the type of content that's going to resonate. But you never know. It could suck. We'll see. So, interesting, interesting. So it actually, this might be an edge case because when you go to ideabrowser.com, it's just showing, basically every day we share a free idea that people can go ahead and steal these ideas and build it. So today's idea is AI agents that find profitable, expired domains. So it's scraped the website and it didn't realize that we have ideas every single day.

We have database of ideas. We have trends. We have this thing called idea agents. It just thinks it's a website for agents for domain. So it didn't do a good job at scraping ideabrowser.com. But I've had this with other LLMs before. It's kind of like a niche because it's kind of like a new type of product. But I'm just going to tell it like, hey, you didn't do a good job scraping ideabrowser.

You only scraped the idea of the day. The idea changes. Today's was the AI agents for domains. Idea Browser is the number one tool for people to find ideas to build and trends to capture. And there's a bunch more. so don't just browse the homepage. So if you do come across, this is not just for GROC4,

but if you do come across something that's wrong, just tell it. Like, don't give up on the search. Like, I could have easily just given up on it and just been like, yeah, you know, GROC4 sucks at content marketing, but it might. It might not be good at all,

but I just want to double-ch- check because the upside of this working is pretty huge. Like if we can figure out a way to generate more content, that means more leads, more attention. And, you know, it's worth it for me just to do that second prompt and see what happens. And you're going to come into issues like this. You know, so, but just treat it like a human being. Okay.

So now, yes, it's figured, it's properly scraped. idea browser.com. And now let's see what happens. Okay, so the content agent, 10 viral ex post ideas. Each with a hook, trend tie in, and the snippet from an idea

browser idea and CTA, designed for threads, polls, visuals. Okay, Bastidei Day AI revolution idea. This is a horrible, horrible. Actually, this is all bad.

Like, so much hashtags, like, this is all bad. The SEO keywords. I mean, I just feel like how many people are searching best E.J business ideas? I just, I don't see it. Okay, so the content agent

wasn't very good, in my opinion. Let me know if you think otherwise. All right. Overall, not so good. I'm going to try one last time. I'm going to say, you didn't nail the content for ideabrowser.com. Just give me five sample tweets that you think would go viral. You are overthinking it.

The audience is people who love startup ideas and who want a playbook for how to implement them. So maybe the prompt wasn't the right prompt, you know. And maybe I just should have done a easier way, you know, just a more simple way of doing it. What I probably should have said actually is look at X posts. And I wonder, see, on Manus, what you could do, which I wish more LMs would have, is that like if it's working, you can still type something. and press, submit. But, you know, you can't do that here. You actually have to stop the content.

So here it's giving, again, it's like all hashtags and stuff like that. I'm going to say one last thing. You know, I like the style of ideas from Akra, Eisenberg. Make it more like that. Look at his X. And let's see. what happens here. This is my last time. I'm going to try this before we move on to the customer feedback analysis agent. So far, not super impressed with the content marketing agent.

And let's just see if it's my fault, if I'm not prompting it correctly, or if it, or if GROC 4 is just not amazing. at this particular piece. Okay, wow. So this is really good. So it's taken my style of how I write startup ideas. And it's, wow, this is really good. And it's created new startup ideas, written it very similar. So it basically looked at my format for how I write startup ideas.

So I had one. Here it is. I just wrote one a couple hours ago. Startup idea for you. AI agents have find expired domain. So I wrote this up pretty quickly. And it started to get a bunch of traction. And Grok went ahead. And Grok basically created new ideas in like the same casual, clear tone. And it did a good job. So if you want to use Grok 4, The lesson is it's really good for X. I need to play with this more. You just do it simple. I like this person's tweets. Go ahead and make me tweets like that. But for your niche. So that's kind of interesting.

And let's move on. All right. So customer feedback agent. This is super helpful because for ideabrowser.com, here, I have some sample feedback. but I don't really want to go through every single one. I kind of just want an agent to run through it and tell me what I need to know.

So the prompt is analyze this batch of customer reviews. You can upload a file or just paste it. As a sentiment analysis agent, categorize themes, quantify NPS score, and suggest product improvements with prioritized roadmaps. Integrate real-time X mentions of my brand for broader insights and forecast my impact on retention. That's pretty crazy.

If that's able to do that, that's pretty crazy. I'm a bit skeptical around, you know, forecast impact on retention. But if it can figure out NPS score, if it can categorize themes, that would be huge. And if it could do it in a way that's not going to take, you know, a long time. Like, could it take under two minutes to go through all these, all those, all this feedback, that would be really, really cool. And it is cool to have the tie in with X. You know, as someone like me, I love X, I use it a lot.

You know, I want, you know, a lot of my world is on X, right? A lot of people talk to me on X. A lot of, you know, it's just for people who are building startups on X, it's, and have a bunch of people interacting on X. It is helpful. Okay. we've gone ahead and categorize the themes. Wow. Okay. So it's the themes are content and idea quality, UI, U.S. and accessibility, performance reliability. Wow. Okay. So it gives you the

sentiment summary. So like on idea quality, people are saying highly positive user, users love data backed ideas. They love trend signals. They love AI agents for validation and pivots. Minor tweaks like videos or summaries were requested. So then someone, you know, it pulled one of the testimonials, Greg's picks, which is like on the database, there's some ideas that I actually, you know, will pick and say they are good. Feels like a cheat code. So this is amazing. Like if you aren't doing this and grabbing feedback and getting feedback and putting into GROC 4, you're missing out because this is such an easy way to win and give you ideas, startup ideas and feature ideas on where to take your business. it goes and gives it an NPS, wow, an NPS of 90, which is crazy. It tells you about what the NPS score is. And it says this is exceptionally high indicating world class loyalty. Like X mentioned support this with users organically promoting ideas.

Yeah, because on X some people say get this into idea browser. So this is holy, some of the. then it gives you a full-on product roadmap of where to take it based on feedback. And then it gives you a forecast. Okay, so based on that, it gives you the forecast impact on retention. No brainer. This might be my favorite part so far. Because it's just, it's on, it's really on point. All right. So the prompt six, the customer feedback analysis agent must, must do. It's worth worth doing. Okay. The next prompt we're going to do is going to be around the

negotiation agent. So prepare me for, let's say, a salary negotiation. Based on this scenario, describe details. I'm currently making 85K per year, but have been stuck making that for three years. with limited equity upside two. And I feel like I have increased output 2x, yet pay has remained the same.

It says role play has both sides using multi-agent mode. Generate scripts, anticipate objections with data from recent deals, search X and web, and suggest win-win tactics to increase my deal value by 50%. So let's see if this works. Things to note here, multi-agent mode, it's probably important to include that in your prompt. Generate scripts, anticipate objections.

And again, we want to say search X to get the most out of GROC 4. 25 seconds, and we've got an answer. earning 85K for three years with no raises you're in a strong position to negotiate especially if you can try your contributions to measurable business impact win-win tactics to increase deal value by 50% focus on total compensation rather than base salary alone to create mutual benefits okay use sites like glass door levels FYI to benchmark your roles market rate performance-based bonuses prepare by scripting your pitch role playing with a and timing the ass post-major win.

That's actually really smart after a project success. That's really, really smart. I'm going to simulate a negotiation between two agents, you and the employer, anticipating common objections like budget constraints. I'm just reading this now. Like, hi, manager, thanks for meeting. I've been reflecting on my three years now where my output is doubled, handling twice the workload.

This is really good. This is really good. So using the multi-agent, I'm sure people are reading this now. Like, this is just, it's really clear, clean. And yeah, I think it's, if you're going to negotiate anything big or you just want someone to, you know, a negotiation partner, basically. Like, it's going to help you get better at negotiation. GROC 4, pretty darn good.

All right, we've got two more. And then we're done here. And we'll make a call if we like GROC or not. forecast emerging trends in, let's say, you know, let's go with productivity apps. For the next 12 months, using real-time X data and web search as an innovation agent, proposed three product features my startup could build, including a feasibility analysis, competitor gaps, and projected revenue impact, use code mode to model a simple growth projection. So thing to note there, code mode, X data,

Be specific around time, you know, 12 months. Be specific around what you want. Three product features, not product features. Sometimes I find the more specific you are with any LLM actually, the better output you get out. You do have to have some guardrails or else, or else these LLMs, man, they got a mind of their own. So let's go ahead and see what,

what's happening here. So it's analyzing the task requirements. The startup specific focus isn't mentioned. So that's interesting. So maybe I wasn't, you know, I shouldn't have been, I should have been a little more niche than just productivity apps. Maybe productivity apps for remote workers or something like that. But you can see it's executing code right now,

which is so cool. So cool that, you know, you can do this. Okay. So a minute and 10 seconds, a bit longer than the others, but let's see what we get. It says, here's a synthesized forecast for July 25 to 26. From recent expert analysis, user sentiments, and market projections. Productivity apps will increasingly incorporate multi-agent systems to handle end-to-end workflows, such as report generation, task automation, and decision-making with minimal human input.

generative and flexible interfaces that's completely correct. These are both correct. Traditional rigid UIs will shift to canvas and node-based designs enabling unrestrained, restricted visual thinking. That's absolutely correct.

Deep integration to existing workflows. This is really good. So it's giving you product features to the canvas-based general task interface. I actually think that that's completely where things are going. And then it gives you a growth projection. This to me is less.

interesting, you know, but this is really good. Proposed features and trend analysis is really good. And you can go ahead and like if you don't have a trend, like go to something like ideabrowser.com, find a trend, put it into grok, get some ideas as well. So the last one we're going to do is see if grok is any good at being a design agent. I'm pretty wary of this.

I'm not sure that it can do it, but let's go ahead and see. Create a logo for my startup, ideabrowser.com with a futures of vibe according elements with circuits and stars. Generate four options in vector cells, rank them by potential brand appeal using marketing trends from X

in the AI startup community. So let's see. I have noticed, you know, with a lot of these LLMs, like they do a pretty good job getting something up, but the copy is off sometimes. Like they spell things wrong. But, oh, actually, I was wrong. Here it's spelled completely right. It did what it asked for. It's actually pretty tight, pretty clean. I asked for, well, it's, well, it's spelled completely right. It's, it did what it asked for, it. It's actually pretty tight, pretty clean. I asked for, well, I asked for four options. It gave me two. So it didn't do that, but it spelled it right, which is really cool. And I do like how it has this edit button right here. So if we said, what do we want to do here, make it more colorful. Let's see what happens. By the way, it was like really quick.

I'm not sure how long that took, but like probably like under 30 seconds to create a logo, which is crazy. let's see if the colorfulness make it a little bit more interesting. And there you have it, folks. That, it broke on the copy there. That actually looks pretty good. But, yeah, it broke on the copy. This is what you're seeing a lot of the time. It happens a lot of time in chat GPT as well.

but I guess if you can continue prompting it, you probably can get something you like. And even this is like pretty good. So it's a good starting point. It's good to create maybe your mood boards and stuff like that. Are you going to use it for your $100 million year of business? I don't know. But for getting ideas out there and iterating,

I do like this edit button. So I'll give it that. The other thing is sort of a bonus thing. I saw this tweet. They just enabled companions on Super Croc. And this is how you can enable it. Companions are basically, I actually haven't used it. Let me pull it up.

I'm going to pull up Grock on my phone. Okay. So I pulled it up on my phone. I clicked enable companions. And let's see what happens. And then you download these companions, which I think are. are just like AI friends.

So I'm going to, I'm downloading one of the companions. And let's just see what that's like. It's kind of like that movie, her, I think. So interesting that, okay, Grock, okay, I don't know if you can see it. But I'm going to go talk to, hello, my AI friend. Hey, cutie. And my black dress, I know the tea.

What's the weather like in Toronto, Canada today? It's another hot, humid day in Toronto. With the temperature at 24 degrees Celsius, feeling like 28 degrees due to the humidity. There's also a heat warning in effect, with poor air quality that might make it tough for some to breathe outside. If you're planning to be out, maybe stick to shaded spots and stay hydrated, babe.

How's the day going in all this heat? So insane that it called me babe. Don't feel good about that at all. Married man over here. So you're thinking about that hot, humid weather in Toronto. So that's, it's kind of weird, but, you know, there you go, Companions Mode on Groch. The last thing I want to cover is I had heard that voice mode on Grok.

really good. So I'm going to go ahead and I'm going to ask it to brainstorm revenue ideas for my startup and let's see if it works. So brainstorm revenue ideas for my startup, ideabrowser.com. It's for ideabrowser.com. I'm assuming it's a platform for browsing or sharing ideas, but correct me if I'm wrong. Here are a few ideas to start. First, you could go with a subscription model. Offer premium features like advanced search filters, collaboration tools, or exclusive access to curated idea collections for a monthly or yearly fee. Maybe a free tier gets users hooked, then they upgrade for more. Okay, so it does sound like more of a human being. I could see myself using someone like that when I'm on a walk or something like that. So TBD on that, but I think

that's got a lot of legs. There you have it. That's GROC4 review. Overall, how am I feeling about it? I think that there's some things which landed really well. Like, let's go back to the list. The market research agent was really cool. The coding, not sure. It looked clean, but I'm not sure if it's like, is that better than cursor or replet?

I'm not sure. So someone, please let me know. The productivity workflow optimization was really cool because I could see myself just asking it for helpful ideas and stuff like that. The fundraising pitch refinement, I thought that was really, really good. I can see myself using this for Dax. The content marketing strategy wasn't good, but then when I asked it specifically around tweets and I gave it a specific person, it did work really well. Customer feedback analysis agent, I'm going to 100% be using this. And I'm thinking to myself, like, how can I do this as an automation, right? So,

every single month, you know, it pulls our feedback, it asks Rock 4, and then it posts to our Slack. So I'm definitely going to be creating that vibe marketing automation. The negotiation script, I think, is really interesting. I just got to make that more part of my workflow. The trend forecasting did a really good job. The branding visuals, I mean, I'm not kind of like, I'll give that like a 6.7 on 10. It was cool that it was so quick and it was pretty clean, but when I actually went and edited it and went deeper into it, it started to break. So hopefully they fixed that soon.

And then, of course, the companions, that was like a bit of a weird experience. It feels like you're kind of like flirting with these AI avatars. but I will say the voice mode was super helpful. So there you have it, folks. In a nutshell, my review, GROC4, AI agents, hot off the press. I want to know what you think. I want to know if you like it, you don't like it. Just use the comment section as a place for us to just share what we're learning,

what we like, what we don't like. Thank you. And I'll see you next time. It's a lot of