Real work at a real startup. A prescription system in development, Meta campaigns going live, and merchandise in production.
RaazMD is a health-tech startup building a men's digital clinic. I joined as a design intern in February 2026, working across brand, campaign, and product design. Most of what I worked on went directly to the founder or CEO for approval, which meant the stakes were real from day one.
The work ranged from a multi-platform prescription system that's now in development, to Meta campaigns, social assets, merchandise going into production, and a full website migration. Different problems, different constraints, same expectation: get it right.
Prescriptions at RaazMD aren't just medication lists. They combine a root cause evaluation with a treatment plan, and the same document needs to work on screen, in an app, and as a printed PDF. Three different contexts with different reading behaviors, different screen sizes, one design that has to hold across all of them.
The founder and my manager assigned this to me with the core requirements and gave me full ownership over the structure and visual direction. I didn't want to get it wrong, so I didn't guess. I explored.
Two directions stayed in the running longest.
The card layout felt intuitive for digital. Each product got its own card, icons, clean spacing. It looked good on a screen. The problem was that it didn't scan well when there were multiple prescriptions, and it fell apart completely on print. Medical information needs to be readable fast, and cards made that harder not easier.
The table layout was less visually exciting but it worked. Dense data, clear columns, easy to scan, readable at any size. For a document that a patient might hold in their hands or open on a phone, clarity wins over aesthetics.
The root cause section went through the same process. The colored pill tags were visually clear but they just showed a list. The categorized approach, grouping deficiencies, comorbidities, and lifestyle factors separately, made the evaluation feel clinically organized. It's the difference between showing information and actually explaining it.
The final system is two pages. Page one covers patient information, root cause assessment categorized into three groups, and the treatment plan in a table layout. Page two is where it gets interesting — a treatment timeline broken into monthly milestones, dietary recommendations, exercise protocols, and important notes. Plus a QR code linking back to the platform.
It's not a prescription. It's a full patient protocol that can live digitally and physically at the same time.
Before handing anything off I made sure the Figma file was clean. Every layer named, every component organized so the dev team could work without having to reverse engineer what I was thinking. I sat with them through the process too — understanding their constraints early meant I wasn't designing things that were impossible to build.
The design has been approved by the founder and handed to the development team. It's currently being implemented across the platform.
I designed banners for every platform the company is active on as part of a unified visual system. LinkedIn, YouTube, X, Facebook — each sized and adapted separately but all coming from the same visual language. The LinkedIn banner is now used by 20+ employees across their personal profiles including the founder. Approved on the first iteration with minor adjustments.
The campaign content and direction came from the content team and founders. My job was taking that narrative and making it work visually on Meta.
The carousel moves across five slides: starting from the problem, through why temporary fixes don't work, into diagnosis clarity, the solution, and a final CTA. Each slide carries the logic forward, not just looks good on its own. The post follows the same visual language but works as a standalone piece. Both are approved and going live.
This one doesn't get shown in portfolios often because most people don't know it exists as a design problem. WhatsApp Business templates have strict constraints — image, text, and CTA buttons in a fixed format, within platform guidelines.
I designed a user engagement template for booking tests, consulting doctors, and getting support. The challenge was making something that feels personal and trustworthy within a format that gives you almost no room to move.
Before making anything I researched what actually works for medical and health content on YouTube — what gets clicked, what builds trust, what reads at small sizes, and specifically what could work for RaazMD's audience and tone.
Then I designed 10+ thumbnails and defined the visual system that governs future ones. Layout rules, contrast requirements, content hierarchy, how text and imagery work together. The thumbnails that come after mine should follow the same logic without needing me to make them.
RaazMD moved from an old website to a new platform. I worked in a team of three across two main areas.
For the homepage and key pages I created graphics aligned with the new brand direction and ongoing campaigns. The two main graphics I worked on are visible on the live site now.
The blog side was a different kind of work. 30+ existing posts needed their visuals updated to match the new brand guidelines. Some images I recreated, some I colour graded, some I replaced entirely when the original didn't make sense for the content. Every post got the logo integrated consistently. It wasn't glamorous work but it mattered — a blog that looks inconsistent undermines everything else the brand is trying to do.
I designed a t-shirt and polo for the brand, working from a brief but handling the full execution. Minimal typography, clean brand presence, something that looks credible when you're wearing it on camera. Both are currently in production and will be worn by the team in videos and internally.
I also contributed to packaging visuals, working on layout and brand consistency for upcoming product packaging.
The table layout isn't exciting. It's right. That distinction matters.
On the prescription system
Working at a startup means your decisions are visible immediately. There's no middle layer between the design and the people who decide if it works. That changes how you think. You stop optimizing for looking good in a review and start optimizing for actually solving the problem.
The prescription system taught me the most. I had full creative ownership which sounds great until you realize it means all the wrong decisions are yours too. Making 10+ iterations wasn't indecision, it was the only way to figure out what actually worked.
Working closely with the dev team changed how I design too. When you understand what's easy and what's painful to build, your decisions get more honest. Clean Figma files aren't just courtesy, they're part of the work.
Next Project