-
-
Homescreen, paste links in the box to generate ingredient lists.
-
Pantry page. Add or delete items from your pantry, or add to shopping list.
-
Shopping list. Mark as bought after buying.
-
Recipe page (after parsing link). Ingredients available glow green checkmark.
-
Recipe page (after parsing link). Ingredients available glow green checkmark.
-
Saved recipes page. Recent links seen below.
-
Settings, set your profile pic, alias, preferred cuisine, preferred categories.
-
Community page. Share your recipes and tips with friends.
-
Ingrid whatsapp account link-up instruction page
-
Ingrid whatsapp kitchen assistant in action!
-
Revenue cat paywall. Made dismisable for testing purposes.
Inspiration
Ingredify was inspired by a real community and a real problem I’ve observed for over a decade.
About 10 years ago, I created Chefs, a WhatsApp group for professional chefs, home cooks and food enthusiasts. Over time, it became a daily space where people shared recipes, cooking tips, kitchen hacks, and — most commonly — ingredient-based questions like:
“This is what I have in my kitchen. What can I make?”
Despite the expertise in the group, answers were fragmented, repetitive, and ephemeral. The same problems kept surfacing: decision fatigue, unnecessary grocery runs, and food waste. Ingredify was built to directly serve this audience — people who cook regularly, care about food, and want practical guidance that respects what they already have.
What it does
The app is intentionally designed for:
Home cooks and busy professionals
Students and budget-conscious users
Anyone who wants faster, smarter cooking decisions
How I built it
Ingredify was built with a product-first mindset, informed by years of real-world cooking conversations from the Chefs whatsapp community. It was built with Flutter/Dart primarily.
I started by defining ingredients as the primary data model, rather than recipes. The core flow was intentionally simple:
- Users input ingredients they already have in their pantry
- Users set their cuisine preference (e.g Asian, African, Intercontinental, Vegan, etc) as well as category for their saved recipes (e.g Pastries, Soups, Stews, Low-Calorie, etc)
- The recommended section on the home page displays recipes based on their cuisine preference, and items available in their pantry.
- Video links can be pasted into the app to generate ingredients list. The items available in the pantry glow green checkmark, items not available in pantry can be added to Shopping List.
- Users can save Recipes and Ingredient Lists and access them in the Recipes Page.
- Users can organise grocery list shopping accordingly on the Shopping List page. Tick off items as you buy.
- In version 8, users can post on the community page, share meals, tips, recipes, and learn from others.
- Users can chat with connect their account with Ingrid, the Ingredify Whatsapp Automation Assistant and ask questions like :
Whats in my pantry? Add rice, eggs, cheese to my pantry. What can I cook under 20 minutes? Add olive oil to my Shopping List.
Ingrid responds based on the user's data in the app, and updates it when prompted accordingly.
AI was used selectively to:
- Connect the Url field to Gemini API to parse the contents on the page and present Ingredient List for the intended recipe.
- To help program Ingrid, the Whatsapp kitchen assistant for Ingredify.
- Also to generate image assets for the project.
On the front end, I focused heavily on UX and visual consistency. Ingredient icons were custom-designed with a unified style and transparent backgrounds to make scanning fast and intuitive. ( I had to purchase a pack of food icons primarily, then manually design the icon of some of the popular food items). Every design and technical choice was guided by one question: Would this make deciding what to cook easier? Would it make it easier to organise recipes, pantry items, and shopping lists?
Challenges I ran into
One of the main challenges was avoiding overengineering. With AI-driven features, it was tempting to add complexity, but doing so risked making the app feel unpredictable or opaque. I had to be intentional about limiting AI usage to areas where it clearly added value.
Another challenge was structuring messy, real-world cooking behavior into a clean system. In the Chefs WhatsApp group, advice is flexible and contextual. Translating that into an app required finding the right balance between structure and flexibility.
Design consistency was also a challenge. Because Ingredify relies heavily on quick visual recognition, even small inconsistencies in icons or layouts were noticeable. Achieving a cohesive, polished UI required careful iteration.
I struggled for a while with the server hosted on Render, to power Ingrid, the Whatsapp assistant. I also had to use Twilio, apply for a whatsapp business account approval with Meta. A lot of hard work went into Ingrid. But I had to ensure it was a feature that was included, because of what Eitan said in the Live session, about how the app needs to be very easy to use for even non tech-savvy individuals. Ingrid solves this problem making the app easy to use for a 10 year old, or an 80 year old, and everyone in-between.
Accomplishments that I am proud of
I am proud of Ingrid! She still needs polishing but it was is wow feature, that would set me apart from the competition and I am glad I was able to pull off MVP level before the deadline. The pantry and grocery shopping list organisation is my personal favourite. Makes life easier.
And also:
- Built an ingredient-first cooking experience that feels intuitive and practical
- Designed a clear, demo-ready flow that communicates value in seconds
- Created a consistent, scalable visual system for ingredient representation
- Used AI in a way that enhances usability without overshadowing user control
- Translated a decade-old community insight into a tangible product
- Users can share food pics without judgement, and learn from others too. Thats a fun feature for me.
Most importantly, Ingredify feels like something people could realistically use every day.
What I learned
This project reinforced the importance of designing for real behavior, not idealized workflows. People don’t cook with perfect recipes or full pantries — and tools should reflect that.
I also learned that AI is most effective when it operates quietly in the background, supporting decisions rather than making them for the user.
Finally, building Ingredify showed how valuable long-term domain familiarity can be. Years of informal conversations in the Chefs community provided insights that no dataset alone could.
I became more proficient with the implementation of Revenuecat in my projects. Spent less time on that this time around :)
Also I gained new knowledge while implementing Ingrid. I am excited for future project I can build using the same Whatsapp Chat Assistant Functionality.
What's next for Ingredify - Pantry Organiser
The next stage of evolution for Ingredify is to fix all minor bugs, connect Eitans Recipes from his recipe books (if he prefers) to the app, and polish up several areas to make it fully production ready and to be monetised. Areas to be fixed ASAP (was not implemented because of time constraint)
- Ensure recipe library is robust(currently has dummy data), has a wide range of cuisines, which will ensure that the recommendation feature (suggesting recipes based on items available in pantry + cuisine preference) works efficiently.
- Notification of comments on community page
- Ability to share recipe links, shopping list links on community page, make them clickable.
- Building Ingrid to be smarter and handle queries more effectively.
- Ensure Ingrid is effectively connected to users account and can read and update user data accordingly. And much more!
To make Ingredify the "Complete Cooking Assistant for professional chefs, home cooks and food enthusiasts"
It's features will allow users to:
- Maintain a persistent pantry inventory
- Track ingredient quantities and expiration dates
- Receive recipe suggestions based on pantry status
- Generate smart grocery lists for missing or expiring items
- Ingrid will be made smarter and more robust, syncing data seamlessly with the Ingredify app.
Longer-term plans include:
- Meal planning based on pantry optimization
- Personalised recommendations based on cooking habits
- Community-driven tips and chef-curated recipes inspired by Chefs
- Development of Community page for inspiration, ideas and sharing kitchen hacks, helping with cooking SOS's. Including more features to make it socially intuitive, like sharing the link to your shopping list so someone else can shop for you, etc. Think Instagram + X + Reddit for food enthusiasts.
The goal is to move from a decision tool to a complete cooking companion — one that helps users waste less, cook smarter, and enjoy the process more.
There is huge potential for Ingredify in the market.
Built With
- dummiesjson
- firebase-console
- flutter/dart
- kotlin
- lottie-files
- objective-c
- render
- swift
- themealdb
- twilio
Log in or sign up for Devpost to join the conversation.