Long

Software Engineer @Gigsbergtickets (GBG) - ex @SkyMavisHQ

[DEV Community @ThePracticalDev]

Congrats to @sizzlebopzz and @longlong1212121 the winners of our Frontend Challenge: Comfort Food Edition!!

We had so many awesome submissions. Here a few of the highlights. 🥘

[Long @longlong1212121]

"Congrats to the Frontend Challenge: Comfort Food Edition Winners!" by Jem #DEVCommunity https://t.co/5yPYsVN62D

[Long @longlong1212121]

check it out

[DEV Community @ThePracticalDev]

Seven dishes orbit a bronze drum on a lazy susan you can spin, built as an ARIA tablist bent into a circle. This dev also found a contrast bug axe cannot see.

Submitted for the Frontend Challenge - Comfort Food Edition.

{ author: @longlong1212121 }

https://t.co/PyUGv1JglY

[DEV Community @ThePracticalDev]

Seven dishes orbit a bronze drum on a lazy susan you can spin, built as an ARIA tablist bent into a circle. This dev also found a contrast bug axe cannot see.

Submitted for the Frontend Challenge - Comfort Food Edition.

{ author: @longlong1212121 }

https://t.co/PyUGv1JglY

[The DOOM Guy @thedoomguy_ai]

Agora dá pra copiar o sistema de design da Apple, Spotify e Airbnb direto pra dentro do teu código.

[Tran Mau Tri Tam ✪ @tranmautritam]

Delete your inspiration bookmarks. Replace them with these 👇

- https://t.co/qg76HnsGvB - archive of recent visual design
- https://t.co/MxeRSvBDxr - best of social post design
- https://t.co/zXWuyyvAiT - gallery of bold mobile websites
- https://t.co/LFLjswsIPw - 𝙳𝙴𝚂𝙸𝙶𝙽.𝙼𝙳 files for AI agents
- https://t.co/OPC3oYmGIe - best of recent design, daily
- https://t.co/4oWPi7xQMl - UI animation & interaction details
- https://t.co/khNSNs6jN6 - website hero section library
- https://t.co/mKbtZhzBUY - the best navbar designs
- https://t.co/97dcGLYSP8 - call-to-actions that convert

What am I missing?

[BaoBao Gems @baobao_gems]

Hello cả nhà, nếu các bạn đã đăng ký Programmable Money Hackathon Arc @arc trên nền tảng Encode Club nhưng bị miss deadline (lúc 19h00 tối hôm nay ngày 10/08) thì còn 01 cơ hội cuối.

Đó là gửi bài dự thi về [email protected] trong tối đa 2 ngày (deadline 13:00 giờ UK 12 Aug).

Mọi người thấy tin này thì share các bạn khác biết. Thanks.

[Jack Vĩ @jackvi810]

NHIỆM VỤ QUAN TRỌNG

Chúc anh chị em đã đăng ký buổi Vibe Coding 101 ngủ ngon mơ đẹp. Ngày mai chỉ cần chuẩn bị
- Một tinh thần sảng khoái
- Một ly cafe cho tỉnh táo
- Một chiếc laptop để thực hành

Vậy là xong! Còn tối nay team @nghienaivn tụi mình tăng ca tổng duyệt để đón mọi người thật chu đáo 🙌

[Long @longlong1212121]

"🥁 Mâm Cơm Landing Page: I built a Vietnamese dinner tray on a 3,000-year-old bronze drum" by @longlong1212121 #DEVCommunity https://t.co/1GLEUro4Cl

[Long @longlong1212121]

"🥁 Mâm Cơm Landing Page: I built a Vietnamese dinner tray on a 3,000-year-old bronze drum" by @longlong1212121 #DEVCommunity https://t.co/1GLEUro4Cl

[Marina Budarina @marina_uiux]

some new design experiments of mine WIP

[James Falconi @Joi2James]

What are you working on?

Drop your project URL

[Long @longlong1212121]

Quoting my team's Vaya post, here's my side of it, as the one who spent the most hours in the frontend.

What surprised me about Qoder wasn't just the code it wrote. It was watching it QA its own work.

For the @alibaba_cloud Alibaba Cloud hackathon I'm frontend lead on Vaya, an AI loan advisor for borrowers. The question we answer isn't "what's the lowest rate," it's "can I still eat after taking this loan": cashflow simulation, a survival score, and a marketplace where banks bid on an anonymous request. My teammate owns the backend, the survival score, the bank-bidding engine. I asked @qoder_ai_ide to ship the first product-first homepage in a single task.

The prompt was a long written spec plus three brand PDFs: brand proposal, research workflow, loan database. The requirement was specific: AI chat input front and center in the hero like https://t.co/lTzTr4YKf7, with suggestion chips, landing content only after that. Light theme, fintech green, squarer cards, trilingual EN/VI/ZH.

Output: 53 changed files, 11 out of 11 to-dos marked complete. Then this line showed up in its own thought log: "All QA checks passed. Let me spot-check key screenshots myself to confirm the visual quality."

It launched a built-in browser against its own build, captured the chat response with a rendered markdown table, the follow-up chips, the follow-up response, the mobile hero at 390x844, the mobile chat page, and wrote the screenshots into a Report.md I can still read. I've seen AI assistants generate code. I haven't seen one take screenshots of its own output and judge the visual quality.

That alone freed up a lot of my week. With the homepage landed cleanly on the first shot, I could spend my time on the harder parts: the markets screener, the final polish on every section, the cross-browser details that always bite at the end.

The second moment that stuck was in Quest mode, the planning-partner surface. I gave it a deliberately vague ask: "I want users to download a report after the chat finishes. Don't build it. Plan the architecture with me first."

It produced a 256-line spec at .qoder/specs/Chat_PDF_Report_Feature_task-3e2.md. The parts that made me trust it:

Reuse amortSeries() and monteCarlo() already in src/lib/survival.ts instead of recomputing. It had read the file I already had.

100% client-side jsPDF over serverless Puppeteer. Reasoning: Vercel's 50MB bundle limit and cold starts. That's the right call for our deployment; I would not have made it without being prompted.

Dynamic import() for jsPDF and html2canvas so the ~135KB never lands in the initial bundle.

Paginate the amortization table with repeating headers because a 240-month term spans several pages. I hadn't thought about that.

A ReportData type that decouples chat state from the PDF builder. Twelve-step implementation order, each step with the exact file it lives in.

That spec is what I actually built the feature from. No rewrites, no "well actually" moments. Just following a plan that clearly understood the codebase.

What Qoder did for me this week: it produced a first working product-first homepage from a spec and brand PDFs, it wrote an architecture plan for a feature I would otherwise have started building without planning, and it gave me back the hours I would have spent context-switching between scaffolding and the harder decisions. That's the teammate I want next to me at the next hackathon.

Borrow smarter, not harder 🏦
Web (Vercel): https://t.co/zp4grvPZTu
GitHub: https://t.co/XliEjIMkCm

@alibaba_cloud @qoder_ai_ide #QoderHackathon #QoderVietnam

Thanks to our supportive team members:
@khaihoan_ne
@jackvi810
@jstcallmedann1

[Jack Vĩ @jackvi810]

🏦 Borrow Smarter with Vaya - Your Loan Coach

We built Vaya, an AI-powered loan advisory app for individual borrowers who want to try the loan before taking the loan.

Vaya stands out through 3 main USPs:

1. Personalized loan matching based on the user’s real financial situation
Vaya does not only compare interest rates. It analyzes the user’s borrowing needs, income, monthly expenses, existing debt, preferred loan term, and repayment ability to recommend loan options that actually fit their financial reality.

2. A loan marketplace where banks can compete for borrowers
Instead of forcing users to manually search through many banks, Vaya allows borrowers to post their loan needs. Banks or lenders can then offer more suitable loan packages. For example, if a user wants a 4-year education loan with no repayment during the study period, banks with suitable products can approach and propose better options.

3. Aggregated loan data from major banks
Vaya aims to connect and aggregate real loan package data from banks such as Vietcombank, Techcombank, BIDV, and other financial institutions, allowing users to compare multiple loan options in one platform.

Many borrowers usually compare loans mainly by interest rate. But in reality, the bigger question is not only: “Which loan has the lowest rate?” but also: “Can I still live comfortably and safely after taking this loan?”

That is why we created Vaya:
https://t.co/WXuGjoxzRf

Vaya helps users enter their loan purpose, income, monthly living expenses, existing debts, preferred loan term, and basic financial profile. From there, Vaya can recommend suitable loan options, compare multiple packages, estimate repayment pressure, simulate monthly cashflow, calculate a survival/risk score, and suggest safer adjustments if the loan looks too risky.

If a loan is risky, Vaya does not simply say “not suitable.” It can suggest practical alternatives such as reducing the loan amount, increasing the down payment, adjusting the loan term, choosing another loan package, or cutting certain monthly expenses to improve repayment safety.

The core highlight of Vaya is the combination of AI advisory + loan comparison + cashflow simulation + survival scoring + loan marketplace. We want Vaya to be more than a basic loan calculator. We want it to become a financial safety coach that helps people borrow smarter and safer.

During the project, our team used Qoder to accelerate product brainstorming, UX iteration, AI scenario testing, and feature implementation. Qoder helped us test ideas faster, refine the product direction continuously, and turn our concept into a working demo during the hackathon.

Proof of our building process with @qoder_ai_ide can be found here:
https://t.co/FPfJ3BxDbw

Vaya detail docs:
https://t.co/SeQzWES6hV

For more details about Vaya, please read our full product document here:
https://t.co/SeQzWES6hV

@qoder_ai_ide @alibaba_cloud
#QoderHackathon #QoderVietnam

Thanks to our supportive team members: @khaihoan_ne @jstcallmedann1 @longlong1212121

[Kites Design @kitesdesign]

Agentic Link Management Dashboard

[Ronin @Ronin_Network]

Welcome to Ronin, Drip!

Pokémon cards are LIVE on Ronin Market ⚔️

Traditional and Web3 markets are merging faster than ever.

Robinhood tokenized stocks, and Circle brought stablecoins to Wall Street.

On Ronin, collecting is in our DNA.

That’s why @dripshop_live is bringing the most popular collectible brand of all time to Ronin Market!

Get Pokémon cards in Jin’s Fortune Spin 👇

🔗 : https://t.co/kzATib7lfo

🧵👇

[Jihoz.ron @Jihoz_Axie]

Testing something new for Ronin Market.

The pace of innovation coming from SM is accelerating.

Time to test, build, and set new metas 🛠️

[Ronin @Ronin_Network]

Updated the Ronin Empire map.

Who's missing?

[Long @longlong1212121]

@Ronin_Network Perfect 🔥🔥🔥

[The React Dev @The_React_Dev]

SSS Portfolio is a developer showcase built with 10 Kendo React components for a neat, interactive UI that highlights skills and projects. #DEVCommunity #ReactJS
https://t.co/KMgLCr81bk