DEV Community: Dmitriy Parhomenko

🎉 Completed AWS Generative AI Applications Specialization!

Dmitriy Parhomenko — Mon, 15 Sep 2025 02:23:44 +0000

I’m really happy to share that I’ve completed the Amazon Web Services (AWS) Generative AI Applications Specialization!

Course Certificates Completed:

AI Fundamentals and the Cloud
AWS Services for AI Solutions
Bringing Ideas to Life Using AI

My Thoughts on the Course

Right from the start, I enjoyed the way the AWS instructors delivered the material. Their upbeat, friendly videos made the course feel less like “studying” and more like having a conversation with people who love what they’re teaching. Big thanks to Alex G., Oksana Hoeckele, and Rafael Lopes for keeping the energy high and the content approachable.

The first lab totally surprised me. I expected a sandboxed simulation, but instead I got a real connection to the Amazon Bedrock platform. For 24 hours, I had live access where I could:

explore the platform,
compare different LLMs,
tweak configurations and see how outputs changed.

That hands-on approach made the concepts click much faster than just theory.

And then there was Amazon Q, which quickly became my favorite AWS service from the course:

In the terminal: it feels like magic. With a single prompt you can generate system commands, spin up folders and files, or drop content into them. All those repetitive steps I usually type out? Done instantly.
In the IDE (like Visual Studio Code): it works more like a coding buddy, similar to Cursor or Windsurf. Amazon Q can suggest code, explain snippets, and even guide you through building out a project. It’s the kind of tool that makes you wonder how you ever worked without it.

One of the coolest things overall was seeing practical use cases. For example: a project that checks free parking spots near your house and pushes notifications straight to your phone. It’s simple, but it shows how these services can connect everyday problems with real solutions.

What I’d Love to See Improved

The course gave me a solid foundation, but I’d love to see:

More end-to-end projects that use AWS AI services in a real application.
Additional labs with longer, real-world scenarios so you can practice building something from start to finish.

Final Reflection

All in all, this specialization was a fantastic starting point for diving into Generative AI with AWS. It gave me both the theory and the hands-on practice I was hoping for.

I’m excited to keep experimenting with these tools, especially Amazon Q and Bedrock, and see how I can bring them into future projects.

👉 Have you tried building with Amazon Bedrock or experimenting with Amazon Q? I’d love to hear what you built and how it went!

You can check out the specialization here: AWS Generative AI Applications Specialization on Coursera

🎉 I Just Wrapped Up the Meta Front-End Developer Specialization on Coursera!

Dmitriy Parhomenko — Fri, 08 Aug 2025 01:32:12 +0000

At first, I joined this specialization out of curiosity — I’d seen it recommended everywhere and wanted to check out its structure.

And wow — it turned out to be a solid journey from beginner to intermediate front-end skills.

About the Specialization

What makes these professional programs on Coursera even cooler is that they’re Foundation for International Business Administration Accreditation (FIBAA) certified and can even count toward academic credits in some universities.

This one had around 212 hours of learning across 9 courses!

What I Liked Most

Short, clear videos that get straight to the point.
Reading materials that go deeper when you want them.
Quizzes to test your knowledge as you go.
Built-in code editor for hands-on practice.
Loved the AI Coach — super helpful for reviewing topics.
The final project ties everything together really well.

A Few Downsides

The course forums felt kind of empty — some people just post random characters to “complete” the task.
No mentor feedback from instructors.
Projects are peer-reviewed — sounds great in theory, but many submissions are empty files or copied content. Feels like some are only there for the certificate, not the learning.

Final Thoughts

All that said, I think it’s a great starting point if you're looking to get into front-end development.

Just make sure to actually build and explore, not just tick boxes. Your growth will come from practicing and creating beyond the course material.

View my certificate: Meta Front-End Developer Specialization

After the Hack: What’s Next for MyMealMind?

Dmitriy Parhomenko — Sun, 27 Jul 2025 22:21:52 +0000

This is a submission for the World's Largest Hackathon Writing Challenge: After the Hack.

It started with a simple idea — “I need a recipe app that doesn’t live in 47 open tabs and random notes.”

Now, here I am post-hackathon with something way bigger than I imagined.

What’s Next?

I’ve got two exciting (and slightly chaotic) plans for MyMealMind:

The All-In Service Plan
- Continue evolving MyMealMind into a full-blown platform.
- Add more community features, pro perks, gamification, and SEO-ready tools.
- Turn it into a go-to hub for food lovers, chefs, and kitchen pros.
The Open Source Template Plan
- Polish the code and docs.
- Launch MyMealMind as a reusable template for anyone who wants to build their own recipe-powered platform.
- Think of it as a digital starter pack for food-tech projects.

And who knows — maybe I’ll end up doing both. (Because sleep is overrated, right?)

What I Learned

Building fast doesn’t mean skipping quality — just means you lean on the right tools (Bolt, you legend).
Supabase + RevenueCat = chef’s kiss for scaling with auth, payments, and gated content.
Design + Dev = magic when you stop fighting with yourself over button colors at 2 AM.

Post-Hack Power-Ups

Sharpened my frontend skills
Got hands-on with Stripe + RevenueCat integration logic
Built real features that scale (and didn’t crash... much)
Joined the world’s coolest dev challenge (shoutout Bolt.new!)

Final Thought

The hackathon might be over, but I’m just getting started. Whether MyMealMind becomes a viral app, a dev boilerplate, or both — I’m excited. And hungry.

Beyond the Code: My Hackathon Journey with MyMealMind at the World's Largest Hackathon

Dmitriy Parhomenko — Sun, 27 Jul 2025 22:06:28 +0000

This is a submission for the World's Largest Hackathon Writing Challenge: Beyond the Code.

When I joined the World's Largest Hackathon, I was thinking: “Cool, I’ll build a little recipe app.” What actually happened? I ended up pulling late-night coding marathons, becoming best friends with Bolt (my AI copilot), and discovering an amazing global community of builders just like me.

What made this hackathon feel like a team sport (even solo):

Bolt.new, my 24/7 coding buddy — From tricky logic to naming variables like “authTokenThatActuallyWorksThisTime”, Bolt had my back. No judgment. Just solutions.
@addyosmani , you're a legend — Your tweet pulled me into this whole wild ride. I owe you coffee.
IRL MVPs — My friends cheered me on like I was in the finals of Iron Chef... except I was cooking up JSX and Tailwind classes.
Devpost + Discord = 💯 vibes — Every announcement felt like a hype squad cheering me on. Even when I broke the build. Again.
"Wait, I just built that?!" moments — Integrating RevenueCat + Supabase + Netlify with edge functions, subscriptions, gated logic… me one month ago wouldn’t believe it.

TL;DR

This hackathon gave me way more than an app. It gave me confidence, growth, and a weird obsession with improving UX.

I shipped something real. I met cool people. And I proved to myself that I could do hard things—with a side of fun and caffeine.

Catch you at the next one 🚀

— Dmitriy

Check out MyMealMind on Devpost

How I Built MyMealMind — A Full Recipe Platform in Just Weeks with Bolt

Dmitriy Parhomenko — Sun, 27 Jul 2025 21:53:44 +0000

This is a submission for the World's Largest Hackathon Writing Challenge: Building with Bolt.

What I Built: MyMealMind

MyMealMind is a full-stack web platform where users can:

Save and organize their recipes (privately or publicly)
Explore the most loved meals from the community
Create their own digital cookbook
Upgrade to a Pro plan to unlock unlimited publishing, chef profile, and more

Live Site: https://mymealmind.online

Devpost: https://devpost.com/software/my-meal-mind-online

I built this solo to solve a real-life need — I was tired of losing recipes in screenshots, tabs, and notes. I wanted something beautiful, scalable, and practical — so I built it.

Tech Stack

Frontend: React, TypeScript, Tailwind CSS
Database & Auth: Supabase (PostgreSQL, RLS, Storage)
Payments: RevenueCat + Stripe
Hosting: Netlify (linked from Bolt)
Automation: Supabase Edge Functions
Platform: Bolt.new

Building with Bolt

Bolt was the engine behind this whole project. I was able to:

Develop quickly with built-in Supabase + Netlify + GitHub setup
- Bolt instantly scaffolded a full-stack project with built-in Supabase and database structure connected
- Netlify deployment was wired automatically, making it easy to preview new commits live
- GitHub repo tracking every commit with clean Git workflows for branching and pushing updates
Instantly preview changes with each commit
- Every change pushed to Git auto-deployed and previewed on Netlify-linked environments
- Seamless branch-based deployments with live URLs
Use Bolt to help with tricky logic and prompts
- Wrote helper functions, Supabase RLS policies, React form validation, conditional rendering
- Drafted UI logic for features like "like system", "favorites filter", and gated access
Keep secrets safe with easy .env handling
- Environment variables managed safely in Bolt, with different keys for test and production
- Clear separation between public and server-only secrets
- Stripe + RevenueCat keys handled cleanly without hardcoding
Project Structure Overview
- .bolt/ — Bolt-specific project data/config used for deployment and project scaffolding
- public/ — Static assets (images, favicon, Open Graph preview images)
- supabase/ — Supabase schema, RLS policies, and Edge Functions
- src/ — Main application codebase:
- components/ — Shared React components (Header, Footer, Sliders)
- context/ — Global providers for likes, auth state, pro status
- hooks/ — Custom reusable hooks (e.g. useProStatus, useAuth)
- pages/ — Main route files for login, register, community, dashboard
- types/ — TypeScript declarations for recipes, users, and plans
- utils/ — Integrations like RevenueCat, Supabase auth, Stripe, email
- App.tsx — Main app layout and route wrapper
- main.tsx — Root React DOM mount
- vite-env.d.ts — Type declarations for Vite environment
- index.html — Main HTML entry (Vite)
- tailwind.config.js — TailwindCSS design system
- postcss.config.js — PostCSS setup for Tailwind
- vite.config.ts — Vite bundler configuration
- tsconfig.*.json — TypeScript base and project configs
- package.json — Dependencies and script definitions
- .env.example — Environment variable template file

RevenueCat + Stripe

As part of the sponsor challenge, I integrated RevenueCat with Stripe’s Web SDK.

Created two plans (monthly and annual)
Synced Stripe products with RevenueCat offerings
Added logic to unlock Pro features via entitlement

When a user upgrades:

They immediately see a Pro badge
They can publish unlimited public recipes
Their plan info updates in the Account section

Challenges

Managing test vs live API keys across RevenueCat + Stripe
Designing a public/private publishing model
Handling delayed email confirmation for login
Real-time entitlement refresh after upgrade

What I Achieved

Community with likes, filters, and chef pages
Private recipe collections
Stripe-powered Pro upgrades
Polished UX across devices
Fast, secure, scalable stack

All shipped solo, while balancing a full-time job.

Reflections

This was my first time using Bolt and RevenueCat — both exceeded my expectations.