Marketer for Startups | I help startup founders develop marketing strategy, execute tactics and scale their businesses

You've tried everything. Reddit, LinkedIn, Product Hunt, communities.
Zero traction. Zero customers. Here's what's actually broken...

After diagnosing 100+ B2B SaaS founders, I discovered 58% have the same execution gaps:
  • They post where their competitors are, not where customers gather
  • They mention their product too early (or too late)
  • They abandon tactics after 5 attempts instead of iterating

I've guided founders from 0 to their first 10 paying customers by fixing these gaps.

Not with more strategies. With better execution.

→ Follow for daily execution fixes
→ Follow me on Medium (https://medium.com/@shamal.r.badhe) where I write in detail why your marketing gets zero traction and how to fix it
→ DM if you're stuck and need immediate diagnosis

Shamal Borole - Badhe

We build AI-powered sales systems that deliver 3-5x ROI within 90 days for B2B founders handling their own sales and companies with established sales teams.

Our systems are designed for two specific scenarios:

For founders handling sales themselves:

- Automate prospecting so you can focus on delivery and growth
- Create intelligent follow-up systems that work while you sleep
- Build qualification frameworks that bring only the best leads to your calendar

For companies with sales teams:

- Transform how your reps spend their time (from 28% selling to 70%+)
- Implement analytics that show exactly what's working and what isn't
- Create systems that scale your best performers' tactics across the entire team

This is for you if:

- You're a founder who needs to optimize your personal sales efforts
- You manage a team that's drowning in administrative sales tasks
- You want systems that deliver predictable results, not just more activity

Ready to build a sales system that actually works? Message "SALES SYSTEM" or email me at riccardo@upliftedconsulting.com

Riccardo Vandra

A friend sent me this photo at 6am with one word.

"Dannato."

For the non-Italians: somewhere between "I'm cooked" and a full existential crisis.

I introduced him to Claude Code. 

Showed him AI employees, skills, agents, compounding context.

What one person can build without hiring anyone.

This is the face of someone who just fell into the rabbit hole.

Pietro Acerbis - you're not coming back out.

AI Workspaces for Service Businesses | Founder @ Uplifted

The standard way to work with Claude Code is text.

You type what you want. It types back. For 90% of tasks, that works.

But some problems are visual.

If you're mapping an automation workflow, you want to see where things branch. Not read a 12-step list.

If you're designing a thumbnail, you want to see the layout. Not describe "move the text left and make the face bigger" in words.

Text can't solve that. So I started building Playgrounds.

Single HTML files that give you a visual interface for any problem.

You interact with the problem visually — add steps, mark pain points, annotate changes.

Then the playground generates a structured prompt that you paste back into Claude Code.

It already knows what you decided. No explaining. No re-describing.

Visual problem in. Structured action out.

And in the video below I go over:

- What Playgrounds are and how to install them
- A live build of an automation workflow mapper
- A thumbnail designer that feeds into AI generation

Link in the first comment.

A big part of the AI Consulting game in the upcoming months will be

How to turn human knowledge into agent knowledge

Here's a thought exercise that I run with my AI clients in every single consultation:

"Imagine your employee would have the same context as you when it comes to doing a specific task - would it be able to execute in the same way as you?"

100% say yes

Then I ask: "What if that employee would be an AI Agent?"

That's when they push back and start saying things like: 

"That's not possible"
"AI cannot do that"

But in the end it's just a context problem

It's all about giving the AI the right context

It's just another engineering problem

AI Transformation Partner for Sales and Marketing Teams

"You've unlocked a beast in me."

Got this message from Matthew Farah.

This came just after the first session we had together where I helped him get a foundation on how to build his AI Workspace

And then he just... kept going.

Claude Code for the logic. 
Supabase for the data. 
Lovable for the dashboards.

Now he's building a full internal intelligence system for his business. By himself.

His laptop is basically a 10-person engineering department at this point.

That's the whole point. 

You learn the system, we build it together, and you walk away with a working setup tailored to your business.

That's what I'm doing with AI Workspace Kickstart

I'm taking 3 more service based businesses this month and helping them build their own AI Workspace

DM me "AI Workspace" if you want in.

Being AI First doesn't mean using ChatGPT

It means building a data foundation so that AI can use that

AI is fancy
Data isn't fancy

But that's the necessary foundation for you to really win

And that's exactly what I've been saying for the past 2 years

First test with Claude Opus 4.6

Using their new agent teams functionality

The system: 

- Creates a team of agent
- Each agent has its own system prompt
- Each agent has a specific context that you can visualize
- Each agent can be prompted

How is this practical for business ??

Think about having a problem

When trying to solve this you generally think about how would you solve it, not how others would

In this case, for the research maybe you focus just on the benchmark

Meanwhile other people in your team are more expert in other things. Such as maybe understanding the comparisons better

This is exactly what the new Opus 4.6 allows you to do

Throw a problem at it and see how it will find multiple angles to solve that

Big times ahead

Always bullish on Anthropic

Everyone building AI tools right now has the same fear.

What happens when Anthropic just... does it themselves?

I had this exact conversation on a call this week. 

Two people deep in Claude Code, building AI workspaces. 

Similar architecture. Similar ideas.

And the honest question was: "Are we wasting our time here?"

Because in 3 months, Anthropic could ship an update that makes everything we built obsolete.

And the answer I keep coming back to:

Yes and no.

Yes: if you're building a generic AI workspace. 
That's a race against platforms and you will lose.

No: if you're building the expertise layer that sits inside the workspace.

The interface will be commoditized. 

The skill files, the business context, the encoded judgment - that can't be shipped in a platform update.

Anthropic builds the operating system. 

You build what runs on it.

That's the only play that makes sense right now.

Over the past years Saas was built for humans

But everything is bound to change

Saas was built using 2 main components over his existence

The backend - the part responsable for executing the actions (calls to the database, the techy stuff)

The frontend - the part that we'd see as humans to "make sense" of what's going to happen

But over the coming years Saas will be build for agents instead

Agents don't need to see an interface like us in order to execute actions, they just run code. 

Let me give you an example:

If I, poor stupid human, need to get a transcript from a call I'd have to log to Fireflies, click the right call and then click on 'copy' to get the full transcript

An agent instead runs a quick comand and that's it

And through the use of Claude Code or ChatGPT that's also what we are doing.

We are asking something (Get me the transcript from the call with X) and the AI Agent executes the code to do that

This means that in the future Saas potentially don't even need a frontend

But this creates a lack of visibility and a lack of control for us humans

We'd freak out if we don't know what the agent is doing

That's why saas in the future will be composed of 2 main elements.

1. The backend or the actions that the agent would execute
2. The frontend will just become an interface in which we can check what the agent did and, if needed adjust some errors that we might find

Most people using ChatGPT are doing single-threaded work.

You ask it to analyze 5 competitors. 

It does them one by one. 

By the time it gets to competitor 5, the context from competitor 1 is mostly gone.

Instead, the goal is to start thinking as an AI orchestrator

Capable of spawning agent teams

Like an AI Research Department for example

Here you can have:

→ A marketing researcher analyzing content strategies
→ A product researcher analyzing features and positioning
→ A pricing researcher comparing models and tiers
→ A team lead coordinating findings across all of them

Each agent runs in parallel and shares findings with the other agents

This is how Claude Code Agent Teams work

And in the video below I go over:

- The 3 different levels of AI Work
- How you can enable Agent Teams for your business tasks
- A live example so you can see how everything works

Link in the first comment (or you can check my Youtube)

Every single AI Project should start with data infrastructure

I know

It’s boring

But that’s a step that cannot be skipped

Sure you can get started building fancy automations 

But if you’re serious you should start with the infrastructure

Because once you have that everything can be built 

Thats not the difficult part

The problem is giving agents the right context

and you can only solve that by putting the right data infrastructure in place

taplio

The Battle

Shamal Borole - Badhe

Riccardo Vandra

Followers

Posts

Engagements

Likes

Performance ScoreRiccardo Vandra's LinkedIn Performance

Ready to grow your LinkedIn presence?

taplio

The Battle

Shamal Borole - Badhe

Riccardo Vandra

Followers

Posts

Engagements

Likes

Performance Score.css-goub6s{font-size:var(--chakra-fontSizes-md);font-weight:500;color:var(--chakra-colors-gray-600);display:block;margin-top:var(--chakra-space-1);}Riccardo Vandra's LinkedIn Performance

Ready to grow your LinkedIn presence?

Performance ScoreRiccardo Vandra's LinkedIn Performance