Joy Chatterjee

DCMN

Specialize in developing end-to-end ML infrastructures and solutions to tackle complex marketing challenges.

Work with and integrate advanced platforms such as Codeium, Cursor AI, ChatGPT, replicate.ai, fal.ai and many more to streamline research and accelerate development. Off and on designing scalable pipelines on native and cloud platforms, to activate real-time predictive models, time series forecasting, customer segmentation, ROI optimization and more.

Lead AI Engineer

- Led the project for multivariate time series forecasting of client campaigns, resulting in significant insights in ROI for TV campaigns.
- Implemented LLM-based chatbots for internal teams, using RAG architectures and context
tuning models.

Senior Data Scientist

The Joy of Data

Product Ops & Data Consultant, with over 10 years of experience in data science and three years in product led growth (PLG). Founder of The Joy of Data, based in Berlin.

For data, collaborative ventures with Mercedes, FedEx, Just Mad as well as AI startups such as brighter AI, DCMN focussing primarily into leveraging data for strategic product led growth.

For product, over the past three years, been a data partner for product teams. Responsible for boosting Product Led Growth (PLG) strategies through the application of data science methodologies and data analytics. Expertise domain includes proficiency in product operations, OKRs, KPI driver trees, cohort analysis, user segmentation and much more.

Founder

ShowHeroes Group

Data Scientist (In House Consulting)

brighter AI

Data Scientist

Data Engineer

Mercedes-Benz Research & Development North America, Inc.

Mavrx Inc.

Working on specific project oriented machine vision tasks to aid the system of smart farming techniques developed and monitored by Mavrx Inc.

Machine Learning Engineer

MARKABLE

Rochester Institute of Technology

Work at the lab mainly revolves around projects in the genre of algorithm development. The algorithms are aimed at solving various problems from the image processing and computer vision communities.

Graduate Research Assistant

Peer Advisor Leader

CESC Ltd.

Worked with a team of four to establish and implement control algorithms to modulate the outlet steam efficiency of the central boiler system in a thermal power plant.

Technical Intern

🌱 Founder, The Joy of Data 🎯 Helping Product Teams Grow With Data Science 🚀 Ex-Mercedes | Lead Data Scientist 🎙️ Speaker, Podcasts, YouTube

🚀 Lead AI Research Engineer | Data Science Expert | Product Coach 🚀

With over 11 years in Data Science and AI, I specialize in developing innovative, Data-Driven and AI solutions that elevate Product performance and drive Business Growth. I bring a unique blend of hands-on technical skills and strategic Product insight, supporting teams to leverage Data for informed decision-making and robust growth strategies.

🔹 Key Highlights:
	•	LLMs & Chatbots: Implemented advanced LLM-based chatbots with RAG architectures, boosting team efficiency and workflow.
	•	Predictive Analytics: Achieved a significant ROI increase on client TV campaigns through multivariate time series forecasting.
	•	MLOps & Automation: Created scalable pipelines with Docker & Kubernetes, ensuring seamless deployment and monitoring.
	•	Patented Innovations: Primary author on two patented ML technologies, focusing on object tracking and anonymization.
	•	Cross-functional Leadership: Successfully lead teams and mentor junior Data Scientists, ensuring collaborative, data-driven decision-making.

🔹 Technical Arsenal:
Python 🐍 | SQL 🗃️ | Big Data Ecosystems | Statistical Modelling 📊 | A/B Testing | Cloud Environments (AWS, Azure)

🎙️ Speaking Engagements :
An avid speaker at prestigious Data Science and AI conferences over the past two years, where he shares his insights and experiences to foster a collaborative and informed Data Science community.

Lead Data Scientist

Pavle Lucic

Senior UI/UX Designer @ Toptal | Prototyping, Branding, UI Development

I am a Senior UI/UX Designer at Toptal, a network of the world's top 3% of software engineering, design, and finance talent. With over 10 years of experience in the design industry, I have a proven track record of creating visually engaging and user-friendly interfaces for in-car navigation systems, SaaS apps, and marketing websites.

My background in business and economics gives me a unique perspective on how to design UX and UI that are not only aesthetically pleasing, but also functional, efficient, and user-friendly. I am passionate about solving complex problems, collaborating with diverse teams, and learning new skills. I am also Adwords Search Certified, which demonstrates my ability to optimize the online presence and performance of my clients. My goal is to help tech companies increase their revenue by improving the user experience of their products.

Senior UI/UX Desiger

Nobody talks about the work behind the work.

So let me.

For every design I ship, there are 30 that died.

Ideas I had to let go.
Sketches nobody saw.

Hours on things nobody thinks about.

Error states.
Empty states.
Loading screens.
And so on and on....

But hardest part isn't design. It's the conversations.

Defending a decision.
Moving back and forth.
Rebuilding after a last-minute pivot.

The final design looks simple. 

That's the point.
But simple was never easy.

If your work feels invisible...

You're probably doing it right.

♻️ Repost for every designer whose work goes unseen.

P.S. What's the hardest part of your work that nobody sees?

Senior UI/UX Product Designer | UX Strategy | Figma

How I USED to make case studies:

Step 1: Open Figma
Step 2: Stare at blank canvas
Step 3: Look for "inspiration"
Step 4: Copy a layout from it
Step 5: Realize my content doesn't fit

Then… ( it’s not finished )

Step 6: Start over
Step 7: Give up again
Step 8: Tell myself "next weekend"

(sound familiar?)

How I make them NOW:

Step 1: Open Claude Code
Step 2: Describe my project
Step 3: Gamma generates the deck
Step 4: I am finally Done with case study

3 minutes vs 3 weeks of procrastination.

Same quality.
Zero stress.

Look…

Hiring managers don't care about your case study's typography.
They care about your thinking.
Show them that instead.

Save this. Try it today.

P.S. Tag a designer who needs this 👇

P.P.S. What's YOUR excuse? (I've heard them all 😇)

Junior designers add design elements. 

But senior designers remove.

One rule that fixed 80% of my designs:

Take your finished screen.
Start deleting things. One by one.

In practice:

That extra icon? Remove.
That extra border? Remove.
That extra shadow? Remove.

Keep going until something breaks. Stop there.

Everything you deleted before that? Never needed.

Took me years to trust subtraction over addition. 
Now it's the first thing I do on every project.

💾 Save this for your next design review.

P.S. What element do you always overuse?

Imagine AI could do 100% of your thinking...

Would you let it?

Now imagine a hiring manager asks: 
"What's an original idea you've had this year?"

Could you answer?

(this is the biggest mindset hack from MIT's brain study)
54 people. 3 groups. Brain scans.

The group that used AI for everything?

Lowest memory. 
Lowest creativity. 
Lowest engagement.

The group that thought without AI? 
Highest across every single measure.

I keep thinking...

Your brain is your greatest asset. 
Give it real challenges. It deserves them.

Challenge your brain → Shelve the shortcuts.

Good luck, thinkers! ❤️ (Repost this too ♻️)

P.S. When's the last time you sat with a problem for 30 minutes before reaching for ChatGPT? Genuinely curious. 👇

Here's the truth about designing intuitive UX...

You first start to design → some users may get it.
You design consistently → a lot of users will get it.
You start to give your 100% → everyone will get it.

But what comes after that stage?

When you add more features → a lot of users will struggle.
When you start to "assume" → most users won't follow.
When you expect them to adapt → most will churn.

Unless...You help them "feel" confident at every step.

You offer a sense of "clarity" in every interaction.
You provide more than functionality → You give them ease.

So what is "intuitive UX" exactly?

It's when a user opens your product for the first time and already knows what to do. No tutorial needed. 

The layout, the labels, the flow...Everything just "makes sense." 

Always remember:
Users may use your app for the features.
But they stay for the experience → the effortless, intuitive one. ❤️

(90% of churned users cite "too confusing" as their reason for leaving)

Help them "flow" → not just "function" .

Repost ♻️ if you agree with this

undefined undefined

Senior UI/UX Designer

Senior UX Designer

UI / UX designer & UI Developer

Visual Designer

UI/UX Designer

English

Serbian

How to choose UX research method as a junior?

Need to know HOW MANY users have the problem → Quantitative.
Need to know WHY the problem exists → Qualitative.

Before choosing qual or quant, ask one question: "What do I need to learn?"

I want to know why users churn → Qualitative
I want to understand user emotions → Qualitative
I want to explore a new feature idea → Qualitative
I want to know what percentage of users churn → Quantitative
I want to count how many users have the problem → Quantitative
I want to measure if a redesign improved task success → Quantitative

If you're picking a method before knowing your question...
You're solving the wrong problem first.

Every junior designer can learn this. It's not as hard as you think.

1 question at a time → 1 method at a time

P.S. Repost this if you find this useful ♻️ and save this guide.

There's  2 types of designers.

The "creative" and "disciplined" designer:

Creative designer:

Always has the wildest idea in the room → sometimes too wild to ship
Portfolio is beautiful → but their project delivery is unpredictable
Pushes boundaries → but sometimes pushes deadlines too
Gets praised in brainstorms → but panics during execution
Gets noticed → but doesn't always get rehired

The "disciplined" designer:

Shows up every day with the same energy → Day 1 or Day 100
Delivers on time → even when the work isn't "groundbreaking"
Communicates progress clearly → no surprises for stakeholders
Portfolio shows shipped work → real products, real outcomes
Gets paid → consistently, repeatedly, and without chasing

(Bonus) 

Type 3: 
The rare designer who has both → They run the industry :)

Creativity is the spark.  Discipline is the engine.

One gets you in the door. 
The other builds your career. ❤️

(Invest 3 seconds into reposting this one, will you?) ♻️

P.S. Which type are you right now? Be honest.

Unpopular opinion: Figma is the last tool you should open.

Not the first.

Here's the process I use for every project:

Phase 1 → Discovery (1 week)
Research. Competitors. Customer Journey.
↳ You can't solve what you don't understand.

Phase 2 → Prototyping (2 weeks)
Architecture. Flows. Wireframes. Testing.
↳ You can't design what you haven't structured.

Phase 3 → Design (5 weeks)
UI Concepts. Final UI. UI Kit. Handoff.
↳ Now you design. Not before.

8 weeks total. 

"But that's too slow."
No. 

Redesigning a product 3 times because you skipped research is slow. 

Arguing with devs because there's no handoff doc is slow. 
Rebuilding flows because nobody tested the wireframe is slow.

Process isn't slow.

Skipping process is slow.

(Save + Repost for others ♻️)

P.S. Do you follow a process or wing it?

I've been studying dark UX patterns for a while now.

Here's everything I see on one Temu product card:

"Deal" badge → urgency without a deadline
"17K+ sold" → social proof replacing analysis
1,180 reviews → volume creating false authority
Fake "original" price → anchoring your perception
Cart icon right next to the price → reducing friction
Product image with dimensions → implying precision
Everything above the fold → no scrolling needed to buy
Percentage shown in orange → framing savings over cost

8 design decisions.

But...0 of them are about helping you choose.
All of them are about helping you click.

Feel free to repost this ♻️ 

But my ask is:
Next time you shop online, count the tricks first. Then decide.

P.S. Which one surprised you (1-8)? Anything I missed? 👇

"Good designers have a natural eye for color."

No they don't.

They have a system.

And most of them use the same one:
60-30-10.

60% of the screen is one neutral 
30% is one supporting tone 
10% is one accent

It's simple a ratio.

You could pick ugly colors and still make a clean UI with this rule.
You could pick beautiful colors and still make a mess without it.

The "eye" people talk about? It's pattern recognition.

And pattern recognition comes from systems.
Not talent.

Start with the system.

♻️ Repost or Save if you're tired of the "natural talent" myth.

P.S. Talent or system what matters more? 👇

Underrated design skill: storytelling.

(most designers skip it entirely)

Not storytelling with words.
Storytelling with screens.

1. Every screen should answer 1 question:
↳ "What does the user need to feel right now?"

2. Your visual hierarchy is your "pacing"
↳ First heading = opening line
↳ Call to action = climax
↳ White space = the pause

3. Onboarding is a 3-act story
↳ Act 1: why should I care?
↳ Act 2: reduce friction, build momentum
↳ Act 3: first moment of value

That's it. 3 rules.

Try this → before opening Figma tomorrow:
Write 3 sentences for your screen.
Beginning. Middle. End.

Can't do it?

The design isn't ready yet.

P.S. Do you design screens or stories?

My non-negotiable rules for writing button copy.

(7 micro-copy rules every junior designer needs to know)

Lead with the outcome e.g. "Get Your Free Guide"
"Sign Up" = vague → "Start for Free" = clear value
Keep it to 2-5 words → Longer than that? Rewrite it.
Never use "Submit" → Nobody wants to "submit" anything
Never write what the system does → Write what the user gets
Kill the # 1 objection inside the button e.g. "Try It, No Credit Card"

Read it out loud → If it sounds like a robot wrote it, it's wrong.

Bonus: 
Never use "Submit" on a button because you ran out of ideas. 

If you can't describe the value in 5 words, you don't understand the feature well enough.

Bad CTAs describe the process. 
Good CTAs describe the outcome = That's the whole game.

(Save + Repost for others ♻️)

P.S. Do you follow UX copy rules?

2 years ago I was handing off Figma files.

Today I push code to GitHub.

Same person. 
Same skills. 

One new tool. AI.

I didn't hire a developer.
I didn't go back to school. 
I didn't learn computer science. 

I just started "building."

Backend? AI. 
Databases? AI. 
The scary stuff? AI.

Design, UX, how it feels?

That's still me.
Always will be.

Turns out, a designer who builds the "whole thing"?
Pretty powerful.

♻️ Repost if you know a designer who needs to hear this.

P.S. Which idea have you been sitting on? Tell me below. 👇

Top negotiators use this hack. It's called "echoing."

(here's how designers can use it on every feedback call)

So how it works?
Repeat the client's own words back to them.

That's why it's called the "Echo" method.

People trust their own words more than yours.
So make sure you use that.

Here's how it looks:

Client says: "I want it clean and minimal." 
Then later: "Add more icons, a banner, and a pattern." 
You say: "You wanted clean and minimal. Which of these fits that?"

Client says: "The homepage should feel premium." 
Then sends a reference that looks like a discount flyer. 
You say: "You said premium. This reference feels promotional. Let's find a better match."

Client says: "I love version 2." 
Two days later: "Actually, let's go back to version 1." 
You say: "You approved version 2 last week. What changed?"

No arguing. 

No "I told you so." 
Just their words "Echoing".

What if the client hasn't said anything specific?

Then quote the brief. 
The approved direction. 
The project goals defined.

Why this works: people respect their own words more than yours.

Use playback → fewer revisions, faster approvals.
Start quoting your clients back to them 😇

P.S. Repost this ( Save ⭐ this and Repost ♻️ )

No one will taught you this at any design workshop.

Your clients used to find you like this:

1. Search Contra ( or other freelance platforms )
2. Check your portfolio
3. Like your vibe
4. Read your bio
5. Send a message

Every step there was a real person deciding with their gut.

Do they like you.
Will they hire you.

But, software is now doing that process. 

Browsing profiles. 
Comparing freelancers. 
Choosing who to hire. 

Without a human in the loop.

Now:
AI agents entered the freelance market. 

Which means...

1. A new buyer category that evaluates without messaging you first
2. Profiles being scanned by software, not just humans
3. Less "let me think about it" - more instant decisions
4. More freelancers who have no idea this is happening
5. A new competitor category that delivers work 24/7

The point I want to make is:

Most freelancers don't fall behind because of skill. 
They fall behind because of awareness. 

I really hope YOU won't. 

(Repost further for others ♻️)

P.S. I created a playbook on how to attract these AI agents. 

P.P.S. Comment “AGENT” and I’ll send it your way. 

( Make sure we’re connected first ❤️ )

I never even dreamed I'd be writing something like this.

As of last week, I have my own design tool.

(If you want to stop reading here, go ahead, because I'm about to get a little emotional )

What funny is:

1. I'm a designer. Not an engineer.

I don't write code. I've never opened a terminal. Every design tool I've ever used was built by someone else. Until now.

2. It took me 2 minutes.

I described what I wanted to Replit. A user flow generator with wireframe screens. It built the whole thing. I sat there in disbelief.

3. It does in seconds what used to take me half a day.

I type in any product idea. Full flow. Every screen. Wireframed. I change things by talking to the sidebar. It rebuilds instantly.

The best part is what I realized:

Any designer can do this. 
Right now. Today.

Audit tools. 
Spec generators. 
Component naming tools. 
Client feedback organizers. 
Whatever your workflow is missing.

Designers don't have to wait anymore.

Not for engineers.
Not for product teams.
Not for someone to "prioritize" our request.

We can build it ourselves now.

That changes everything. 
And I mean everything.

Link to try Replit in the comments. 👇

P.S. What's YOUR "tool that doesn't exist"? Drop it below. I'm genuinely curious.

Replit

Every app free trial plan:

Click "Create" ➝ Upgrade to Pro
Click "Export" ➝ Upgrade to Pro
Click "Share" ➝ Upgrade to Pro

After 13 years of product design, I seen this "kill" good products over and over.

The fix?

One feature. 
Fully unlocked. 
Before the paywall. 

Do → stay → upgrade.

Let users "do" something. Not "see" something.
That's the whole UX strategy. 

Written in 3 words: value before money.

♻️ Repost if a product designer needs this.

P.S. What app "lost" you because the free plan was useless?

How to make any product usable without a single design element

(I've done this 3 times already)

1. Take your highest-traffic page
2. Strip every CSS style from it
3. Remove all images and icons
4. Remove all colors and gradients
5. Keep only the HTML and the words

Then:

Watch if they can still convert.
Watch if users can still navigate.
Watch if they understand the flow.

Measure completion rates

Compare to the designed version

Plain HTML + clear copy should be your baseline test. 

This is not a design hack. It's a clarity hack.

taplio

The Battle

Joy Chatterjee

Pavle Lucic

Followers

Posts

Engagements

Likes

Performance ScorePavle Lucic's LinkedIn Performance

Ready to grow your LinkedIn presence?

taplio

The Battle

Joy Chatterjee

Pavle Lucic

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);}Pavle Lucic's LinkedIn Performance

Ready to grow your LinkedIn presence?

Performance ScorePavle Lucic's LinkedIn Performance