DEV Community: Elfreda

The Current State of AI: How It’s Changing Our Lives

Elfreda — Tue, 08 Jul 2025 09:53:09 +0000

AI has made great progress and is now adding value to many parts of our everyday lives. Here’s a look at some key areas where it’s having an impact:

For Everyday Use:
1、Coding: Is shifting from writing every line to prompting an AI partner. The goal is becoming less about typing code and more about giving high-level directions to AIs like GitHub Copilot or Claude.
2、Research: AI agents like Perplexity and the latest ChatGPT can now perform deep research online, consolidating info way more effectively than a simple Google search.

Creative & Productivity AI:
1、Images:
It’s no longer just about generating random pictures. Tools like Midjourney and DALL-E 3、DeepImg offer photorealistic results and more control, moving from “AI generation” to “AI Photoshop.” Think of the viral “Pixar-style moms” or “historical figures taking selfies” trends.
2、Video:
AI video is getting incredibly realistic and is a new cheat code for viral content. Google’s Veo demo of a knife slicing through fruit is a perfect example.
3、All-in-One AI Platforms:
We’re seeing the rise of tools that bundle everything into one place. For example, assistants like ChatGOT integrate top models (GPT-4o, Gemini, DeepSeek) with a full suite of features for chat, writing, PDF, slide , and image.

Other Cool Stuff:

Google’s NotebookLM can turn your notes into a custom podcast.
We’re starting to see AI-cloned influencers on livestreams. Is the future being “on camera” without actually being there?
AI is getting better at math but still can’t invent new ways to solve complex proofs.

On the Frontier:
Google’s AlphaEvolve shows how AI is accelerating science itself:
It broke a 50-year-old record for a core computing algorithm.
It improved the solution to a classic math puzzle (“the kissing number problem”).
On many hard science problems, it now solves them as well as—or even better than—the best human experts.

I Don’t Understand: Why Do So Many Programmers Not Like Using AI to Assist in Coding?

Elfreda — Thu, 03 Jul 2025 09:44:08 +0000

Hey everyone, let’s talk about an interesting phenomenon:

I’ve noticed quite a few programmers around me seem resistant to using AI coding assistants (like Cursor).

I’ve asked a few of them, and their reasons are generally something like:

“The generated code is often junk; it takes too long to fix, so I’d rather code it myself.”
“If I keep relying on it, I’m worried about losing my coding skills.”
“The prompts are confusing, and it’s just easier to look it up on Stack Overflow.”
“Sometimes the model is helpful, but other times it’s totally off—it’s inconsistent.” 👈 I can relate to this point.

But recently, I tried a tool called ChatGOT (fun name, right?), and it seems to address several of my pain points:

Multiple Models: I can switch between different models like GPT-4o, DeepSeek, and Gemini. I can see which one performs best on the same question, which improves code quality a lot. I don’t have to worry about one model suddenly going offline.

Custom AI Bots: I can create an AI assistant tailored to my coding style! By feeding it my project standards, libraries, and naming conventions, the generated code aligns closely with my preferences, which means fewer major changes. No more long prompts every time I write.

Bonus: I can upload requirement documents, and it can quickly summarize or generate a presentation (AI Slides)—great for last-minute meeting prep.

So I’m really curious to hear your thoughts:

What’s your biggest reason for resisting AI?
Is it because you find it inconvenient, or are you worried about being replaced?
Would concepts like ChatGOT’s customizable assistant and multiple models convince you to use AI? Or do you still believe AI-generated code just isn’t good enough?

Just genuinely curious and looking to exchange thoughts!

Which AI is the strongest at writing code?

Elfreda — Wed, 25 Jun 2025 09:53:46 +0000

Recently, I systematically tested the coding abilities of various models on a platform like ChatGOT that combines multiple AIs.

Here are my test results:

Claude 4 Sonnet is clearly leading in agent capabilities. When given clear requirements, it can write documentation, create tests, and write code.

It can also debug, search for documentation, and correct code, continuously working for hours until all tests pass. In this aspect, Claude is unbeatable; GPT and the entire Gemini lineup cannot compete.

When it comes to data structures and algorithms—essentially “hard intelligence”—the O3 series excels the most. Only O3 consistently provides optimal algorithms and architectural solutions.

Claude often comes up with suboptimal algorithms. However, while it has fewer high-level errors, it tends to have a lot of low-level mistakes.

Gemini’s hard intelligence is slightly weaker than O3 but considerably stronger than Claude 4. In terms of following instructions, Gemini is slightly better than GPT, and while it may be the most powerful overall, it is rarely utilized to its full potential.

Most of the time, I use O3 as a professional architect and algorithm engineer to tackle the most challenging parts, while Claude acts as a junior programmer, gradually filling in various details and steadily working in the background.

Gemini’s strong instruction comprehension and extensive knowledge, albeit with some hallucinations, are mostly useful in conversational modes.

Why hasn’t this wave of AI led to a significant increase in job opportunities?

Elfreda — Thu, 19 Jun 2025 09:27:28 +0000

It seems that to become an algorithm engineer, you typically need a master’s degree from top institutions, and even a bachelor’s from a prestigious university isn’t enough.
I’m curious where all the incoming funding is being allocated in terms of job distribution.
As someone involved in AI through my product ChatGOT, I’d love to understand these dynamics better.

My New Time-Saving Formula for Daily Life

Elfreda — Wed, 28 May 2025 09:47:32 +0000

Last week, while helping my kid edit an essay at midnight, I thought: “If AI can free me from these chores, why am I still doing everything manually?” As an efficiency nerd, I’ve tested AI solutions for real-life tasks lately and found some surprisingly simple hacks👇

Scenario 1: From 2 Hours to 5 Minutes for a Last-Minute PPT
A friend urgently needed a proposal slide and sent me a market analysis PDF. No copy-paste nightmares:
1️⃣ Upload the PDF to ChatGOT
2️⃣ AI extracts key sections like “industry trends” and “competitor data”
3️⃣ Pick a minimalist dark blue template → Generate PPT → Send to my friend for final tweaks

My lazy hack: Let AI filter information instead of digging through pages manually.

Scenario 2: Multilingual Emails? No More Panic
When my German supplier sent a technical agreement full of jargon:
√ Compared translations: Google Translate for simplicity vs. DeepL for business tone
√ AI highlighted deadlines and penalty clauses → Turned them into negotiation notes

Fun discovery: Translation models have hidden strengths, like choosing coffee beans for different flavors.

Scenario 3: When Your Kid Demands a “Pop Art Dinosaur Poster”
Saved by AI during a homework crisis:
1️⃣ Typed: “Colorful dinosaur silhouette, Mondrian-style grid background, text: ‘Jurassic Math Competition’”
2️⃣ Adjusted colors (switched from neon to muted tones)
3️⃣ Printed it → Earned “Cool Mom” points

Why Didn’t I Try This Sooner?
▶ Focus on decisions, not busywork: Let AI handle facts; save your brain for creativity.
▶ Skip tool overload: No need to master GPT vs. Claude—switch models like Instagram filters.
▶ Embrace “good enough”: Polish critical docs manually, but let AI handle daily tasks.

(Bonus laugh: When translating a Japanese menu, AI turned “moon-viewing udon” into “udon seen by the moon.” We still ordered it anyway!)

Got your own AI rescue story? Share below—let’s swap lazy genius tricks!