In partnership with
This Deloitte Fast 500 Honoree Is Eyeing 7X Growth
On 55,000+ screens in the US, a new cutting-edge AI platform is modernizing the guest experience across industries.
And the company behind it, Edison Interactive, is just getting started. They’ve even opened a new investment opportunity with plans to grow its footprint from 55,000 to 400,000 screens.
Edison’s AI-powered entertainment platform modernizes screens in venues like golf courses and hotels, helping the company earn Deloitte Fast 500 Honoree. Already deployed at premier locations like TPC, Bethpage, and Caesars Entertainment, it opens new revenue streams for operators and Edison. In fact, they’ve already earned $60M to date.
And to grow its footprint 7X, Edison is doing more than just expanding in golf and hospitality. Verizon is partnering with Edison to bring their platform to professional sports stadiums. From there, airplanes, cruise ships, trains, and more await.
This is a paid advertisement for Edison Interactive Regulation CF offering. Please read the offering circular at https://invest.edisoninteractive.com/
Every week my inbox contains at least one message from someone asking a version of the same question: which AI tools are actually worth using?
It is a fair question buried under an enormous amount of noise. The AI tool landscape right now looks like the app store circa 2011, except every app promises to replace your entire team and make your business run itself. The demos are slick. The landing pages are convincing. The actual day-to-day usefulness is often a different story.
I have been building AI-integrated systems for over a decade. Before The Wealth Grid, I spent years developing automated trading systems that used machine learning models to process market data in real time. I have evaluated hundreds of tools, built integrations with dozens of them, and abandoned most of them after the initial enthusiasm wore off.
What follows is the opposite of a roundup article. It is the tools that survived. The ones that are in my workflow right now, that I would notice immediately if they disappeared, and that I would recommend without hesitation to someone building a serious operation. I have included the honest caveats because this newsletter does not do hype.
The Evaluation Filter
Before I get into specific tools, here is the filter I use. Every AI tool I evaluate has to pass three tests before it earns a permanent place in the stack:
Does it remove me from a low-leverage task? Not just help me do the task faster. Remove me from it entirely, or reduce my involvement to a final review.
Does it improve the quality of the output beyond what I would produce manually? A tool that produces adequate output at speed is useful. A tool that produces better output at speed is a keeper.
Does it connect to the systems I already use, or does it require me to maintain a separate workflow? Isolated tools that live in their own universe have a high exit rate in my operation.
Tools that clear all three tests make the permanent stack. Tools that only clear one or two are probationary. Everything else gets cut, regardless of how impressive the demo was.
Galaxy.ai: The AI Command Center
I am going to lead with this one because it changed my thinking about how to structure an AI workflow. Galaxy.ai (https://galaxy.ai/?ref=danr2) is not a single AI model with a specific strength. It is a platform that aggregates access to virtually every major language model, including the current versions of GPT-4, Claude, Gemini, and others, under a single interface.
The practical implication of that is significant. Different models have different strengths. GPT-4 handles certain types of structured reasoning particularly well. Claude tends to produce more nuanced long-form content. Gemini has specific advantages with certain types of data analysis. Running separate subscriptions to access the best model for each task type is expensive, logistically annoying, and creates a fragmented workflow.
Galaxy.ai consolidates all of that. You open one dashboard, route different task types to the model that handles them best, and you are working from a single interface instead of tabbing between four different browser windows.
Where the platform gets genuinely powerful is in the automation layer. You can build multi-step AI workflows within the platform, sequences that chain together model outputs, transform them, and route them to other tools, without writing API calls or hiring a developer. For a solo operator or small team, that capability removes a meaningful technical barrier.
My primary use cases: first draft content creation, research synthesis from long-form sources, initial client deliverable generation, and competitive analysis. The time savings against manually working through individual model interfaces is real and consistent.
Need a break from high interest?
The right card can give you time—and real savings. Take advantage of 0% intro APR on purchases for an extended period and 0% intro APR on balance transfers for up to 21 months, so you can focus on paying down your balance without added interest.
Make.com: The Connective Tissue
I have covered Make.com in this newsletter before, and I will keep covering it as long as it remains the best visual automation platform available, which as of today it does.
The reason Make.com belongs in the AI stack conversation specifically is that AI tools are, by design, isolated. A language model produces output. That output sits in a chat window until you manually copy it somewhere. Without automation infrastructure connecting your AI tools to your existing systems, you have not actually integrated AI into your workflow. You have just added another tab to your browser.
Make.com is the connective tissue that turns isolated AI outputs into integrated business processes. A few examples from my actual workflows:
A new lead submits a contact form. Make.com sends the submission to an AI model via API, which generates a personalized response based on their specific situation. The response goes back through Make.com and is drafted in my email system, ready for a final human review before sending. Total time from form submission to drafted response: under ninety seconds.
A client completes a weekly check-in survey. Make.com routes the responses to an AI model that summarizes the key points and flags any items that need attention. The summary lands in my project management tool with appropriate context and any urgent items tagged for same-day review.
A new piece of content is approved. Make.com distributes it to my email platform, LinkedIn, and a scheduled social queue, formatted appropriately for each channel, automatically.
None of these workflows required a developer. All of them were built using Make.com's visual interface in a few focused hours. Each one now runs indefinitely without my involvement.
Fathom.video: Meeting Intelligence
If you are still manually taking notes in meetings, you are spending cognitive energy on transcription that should be spent on listening, thinking, and responding. Fathom (https://fathom.video/invite/c-kq_A) handles the transcription problem completely and then goes considerably further.
The tool records and transcribes video calls with high accuracy, which alone would justify the subscription. But the feature that actually changed my workflow is the meeting summary. After every call, Fathom produces a structured summary of what was discussed, what was decided, and what next steps were committed to. That summary is ready within a few minutes of the call ending.
The action item extraction is particularly useful. It does not just note that something was discussed. It extracts specific commitments, who made them, and the associated deadline if one was mentioned. Those items go directly into my task management system via a Make.com integration. I no longer do anything with meeting notes manually.
The longer-term benefit that I did not anticipate: Fathom creates a searchable library of every client conversation. When a client references something we discussed six months ago, I can find it in thirty seconds. When I need to understand the context of a project I have not touched in a while, I can review the relevant call summaries in ten minutes instead of the forty-five it used to take to reconstruct from email threads and rough notes.
Fathom eliminates an entire category of administrative work from client management. The time savings in the first month alone will likely pay for a full year of the subscription.
Rize.io: Time Intelligence
Here is the tool on this list that is the most uncomfortable to use, which is also what makes it the most valuable. Rize (https://rize.io?code=82B5DE&utm_source=refer&name=Dan) runs silently in the background of your computer and tracks exactly how you are spending your working hours. Not how you think you are spending them. How you actually are.
It categorizes your computer activity by type: deep work, communication, administrative tasks, meetings, breaks. It shows you your focus session length distribution and tells you at what times of day your focus is highest and lowest. It produces a weekly report that is either validating or deeply uncomfortable, depending on how well your self-perception matches your actual behavior.
When I started using Rize six months ago, I had a confident narrative about how I spent my time. I believed I was in deep work for four to five hours each day and spending a couple of manageable hours on communications and administration.
The data told a different story. My actual deep work time was averaging about two and a half hours. My administrative and communication time was closer to three hours. And I was context-switching so frequently during what I thought were focus blocks that the quality of those blocks was significantly lower than I assumed.
That data was uncomfortable. It was also exactly what I needed. I restructured my schedule, eliminated several meeting types, and moved administrative processing to a single afternoon block. Within six weeks, my verified deep work time was four hours daily. The work that gets done in those hours is materially different from what I was producing before.
Rize passes all three tests in my evaluation filter with room to spare. It removes me from the task of manually tracking time, produces significantly better output quality through behavioral change, and integrates with my calendar to provide context for what I was actually working on during each block.
Littlebird.ai and Clay.earth: The Probationary List
I want to be honest about tools I am currently testing rather than only discussing what has already proven itself.
Littlebird.ai (https://littlebird.ai/download?utm_medium=referral&source=invite_link&referralcode=ZGFuQHBpbm5hY2xlbWFzdGVycy5jb20=) is a social listening tool with AI-assisted content ideation built in. The concept is strong: it monitors relevant conversations across platforms, surfaces emerging topics in your space, and helps generate content angles based on what is actually resonating right now rather than your best guess. I have been using it for about six weeks. The signal quality is solid. I am still evaluating whether it earns a permanent spot.
Clay.earth (https://clay.earth/?via=dan-kaufman) is relationship intelligence software. It connects to your email, calendar, and other data sources to build an automatically maintained record of your professional relationships, including when you last spoke, what you discussed, and whom you might be losing touch with. For anyone managing a large network of clients, partners, and prospects, the concept solves a real problem. Still in early testing on my end.
Both of these are worth a look if you are actively building in those areas. I will report back with a more informed view once I have more data.
How the Stack Works Together
The individual tools matter less than how they integrate. Here is a snapshot of how these tools function as a cohesive system in a typical week:
Galaxy.ai produces content drafts, research summaries, and initial versions of client deliverables throughout the week.
Make.com routes those outputs to the right destinations, connects data across platforms, and keeps the automated workflows running without supervision.
Fathom captures every client and team call, extracts action items, and feeds them into the project management system automatically.
Rize monitors where my actual attention is going and flags when my productivity patterns are drifting from the target structure.
The division of labor is clean: the AI tools handle the cognitive heavy lifting in their respective domains. Make.com handles the routing and triggering. Rize keeps me honest about whether the whole system is serving me as designed.
The goal is a workflow where the machines handle the repetitive, the mechanical, and the administrative, and I handle the relational, the strategic, and the creative. Every tool on this list was added because it moves the division of labor closer to that ideal.
Want the full AI stack breakdown with specific setup recommendations for each tool? Reply with AISTACK and I will send you the complete guide.
Until next time, build the system.
Alex Rivera
Wealth Architect, The Wealth Grid