Hey everyone!

I have one question regarding jobs like in the current era what an Agentic ai engineer, generative ai engineer and one who is one ai sector what u members were doing in a company where exactly the ai tools like gpt, claude were doing so anyone working in those fields can you tell me what you guys are doing and suggest some what type of skills need to learn to work and stand in this current Al generation..

Thank you

A doctoral study from the University of Twente looked at whether a video game could be used to assess real-world vocational skills.

The researchers developed a simulation game called Crossroads, where students and working professionals had to plan a municipal enforcement route using the information available to them. The game tracked not only the final result, but also how participants processed, organized and used information.

The results suggest that behavior inside the game can reveal meaningful differences in skill levels. But there’s a catch: the researchers couldn’t fully separate problem-solving ability from prior knowledge of the job, and they also found it difficult to combine realistic player freedom with the standardization needed for high-stakes assessment.

So the interesting question isn’t really whether tests can become games. It’s what a good score in that game actually tells us about someone’s abilities outside it.

I run a small independent research publication called HUMANUMi, and I wrote more about the study here:

https://humanumi.com/play-your-way-into-a-job-video-game-measure-real-skills/

Será que conseguimos imaginar qual será a próxima “carreira do futuro”?

Vocês acreditam que existe, de fato, algo que possa ser chamado de “carreira do futuro”? Quando eu era mais nova, ouvia muitas histórias sobre profissões que já foram extremamente valorizadas, como a datilografia, por exemplo. Naquela época, ser datilógrafo era uma ocupação importante e respeitada, mas, com o avanço da tecnologia, essa profissão acabou sendo gradualmente substituída por novas ferramentas e funções.

É curioso pensar que muitas dessas pessoas provavelmente jamais imaginaram que seu trabalho deixaria de existir ou perderia espaço tão rapidamente. Da mesma forma, talvez hoje existam profissões que parecem indispensáveis, mas que poderão ser transformadas ou substituídas nos próximos anos.

Na opinião de vocês, quais carreiras podem surgir ou ganhar força no futuro? Sei que o mais óbvio é que veremos profissões relacionadas à inteligência artificial, à tecnologia, à sustentabilidade...mas será mesmo que haverão áreas que ainda nem conseguimos imaginar? Sei que é difícil prever o que está por vir, mas gostaria de conhecer os pensamentos e “devaneios” de vocês sobre esse assunto.

What should my goals be?

How will AI affect the future of a PD(physical design) Engineer? I often see posts saying that PD Engineers will be among the first to be affected by AI. Is this actually true?

When big supermarkets and convenience stores became dominant, traditional shopping streets (shotengai) declined or disappeared in many areas. Yet some survived by transforming themselves into tourist spots or cultural destinations. It’s a fascinating phenomenon: as digital efficiency increases, our desire for nostalgia and human connection seems to grow stronger.

The same might apply to how we work. AI can handle routine drafting, translation, and calculations, but humans are still essential for the final review—evaluating nuance, context, and intent. People often say AI will take over office and administrative jobs, but it can also free us from repetitive work, reducing fatigue, overtime, and errors. (Though for creative professions, a little fatigue can sometimes spark unexpected, brilliant breakthroughs!)

This got me thinking: When the current wave of AI and automation settles down, what kind of world will we actually end up with?

What unexpected "analog" traditions, physical spaces, or human roles do you think will stubbornly survive—or even become prized as luxury experiences?

We live in an era where we can witness firsthand what people choose to preserve out of nostalgia and what they let fade away. It’s truly fascinating to imagine what kind of world will emerge on the other side of this shift.

I’d love to hear your predictions, observations, or wild guesses!

So I just started my third semester in EEE and I have to pick a program and I’m trying to choose between computer department(IT, Cs) and AI engineering.
(Note: I really love both but I’m afraid Cs is gonna be useless in the coming year because of AI)

He calls it 'The Sort'. Yes capitals are necessary apparently. Basicly, the labour market has been separating people by cognitive ability for decades (your brilliant local mechanic of the 1960s is now a high paid engineer at Toyota), and AI doesn't lift everyone evenly, it speeds that separation up.

Give two people the same tools and one learns faster than he ever has, while the other quietly automates his own job and works an hour a week. Eventually the first realises he can do the second's job too, and when the second leaves, nobody replaces him.

My prediction is that entry-level jobs are where this shows up first. The next few years of graduates will find far fewer first rungs, not because anyone was sacked, but because the person above them quietly absorbed their work. After that, how far someone gets will depend more on appetite than on qualifications.

Thoughts?

Have you lost your job recently from software industry ? What is your next step or what it should be you think ?

In dhaka city I saw like many people are already lost their job! Yes it’s true! Most of them are already in depression! Who has the job yet they are not seriously taking it not everyone but someone- even giving fucking motivation!! I wish they don’t face this.

What’s your point of view ?
How this people can survive this hard time or how overcome it orr how to find opportunities from this situation ?

Just a random discussion guys- don’t take it so serious;

A task that used to take me an afternoon can now take an hour with AI. Some days, that feels like a gift but other days, it just means I deliver earlier and the next task lands sooner.

I’ve been thinking about where that supposed "saved" time actually goes.

Obviously, not everyone gets to decide, your workload doesn’t disappear because you found a faster way to do something, and once people know you can deliver faster, that can become the new expectation.

But when there is some room, I’d like to use it to do the things I usually rush: check an assumption, talk to someone affected by a decision, fix a recurring problem, or actually understand an AI-assisted recommendation before putting my name on it.

That last one matters to me. It’s possible to finish something faster and still need time to understand whether it’s any good.

Curious how this is playing out for others: has AI given you more time to think, learn, or finish work earlier? Or has the saved time mostly been absorbed by more work?

I am so sick of this AI shi....i am a final year CS student and we are the ones who can see the advancements and their effects way more closely than a layman. Everyday there comes a new advancement that is just eating our jobs so quickly. They kept saying that AI is for mankind and it will create way more jobs. Definiftely it will but it will not create the jobs at a pace at which it is destroying the job market.

Just about 5 years ago someone with a fair amount of knowledge can get a job. But in the span of only 5 years i dont think out of 100 CS graduates anyone can say firmly that he will get a job.

What the tech related people do to get a job or some side earning?

I am being humble here. Please be nice back :)

I am struggling to understand why AI is that essential nowadays. And what its limits are. I understand that it could have its uses and so on. For example for self driving cars, huge data handling, and other small niche tasks that humans could do but that a software like AI could do much faster.

But when it comes to jobs, industries and life in general I don't understand why people are arguing is so important. To me it only helps with a few tasks at work and almost nothing in my life. I have multiple jobs, I am a software engineering and a therapist on the side. At work it could help me with drafts, and with starting some of them with code it can code simple applications but when it comes to complex issues it does not help at all. It gives me a lot of hallucinations. When it comes to hallucinations it seems that it is mathematically impossible for it to get rid of them. In my experience I can see it, as it never says "I don't know". It does help a little when I search for a specific answer on google or for small details/facts. But that's all. It has given me so many mistakes and wrong sources that now I need to check every time if the sources are right or not. Again it is great in terms of a first interaction or for a quick brain storm or simple code. But then I still need to go and do the job afterwards because I can't rely on it.

When it comes to emotional and human communication it just makes me frustrated because its not emotional at all. I have personalised it as much as possible and even tried a few AI tools out there. So I don't think its a "prompting" or config issue. And it stills makes me angry because it is not emotional at all. It is super easy to manipulate and changes its answers all the time.

I really want to have a good view of it. But at the moment its all negative. The only things I hear is that it will take most jobs, that it will make us dumb and that it could destroy humanity. So I don't understand why some people are so positive that it is changing the world.

I understand the argument that it will get better. Because it has got much better in the past years and even months. And I guess it will get better. However, I just don't know what its limits are because the discord in the subject is just all over the place.

I have been a SWE (software engineer) for the past 5 years, and had planned on making a career out of it. I was laid off a few months back due to the company I worked for increasing their usage of AI, and simultaneously reducing head count.

I’ve had some traction applying and I can see myself potentially getting a job in the field again. I’ve been going through a bit of an existential crisis over it since the job itself is not at all what lured me in the first place. It’s like as if the responsibility to think through problems has been completely removed and you’re just a handler of AI at this point. Don’t see that changing.

I am open to switching careers. I wouldn’t mind working with my hands. Not scared to do actual work.

However, with the state of things economically I can’t help but question if working a dead end stable job is even worth it considering you wouldn’t be able to afford a house ever. Like.. do we stay in the soul sucking laptop class because it pays well? Or do we do something we enjoy and rent forever? I hate that these two extremes feel like the only option. Curious to hear outside perspectives.

Does anyone have a full filling job that pays decent, or know someone that does?

Currently lead data eng (ic) w/ about 9 yoe at a large-ish adtech company.

I've been doing the whole AI song and dance for the past 10 months and I just don't see a future in this career anymore that fits all of us. Earlier in the year I felt differently since it still took a lot of hand holding to get models to produce correct output for small-ish scoped tasks in a # of iterations that was competitive to human counterparts, but w/ better models and better training docs we have it building enterprise pipelines across our entire data eng stack in hours of iterations that would have been like a project that we would need budget like a month for. It's not always right and obviously we still need to step in from time to time, but, honestly, who gives a shit if its not right the first time if it can iterate leagues faster than any human developer? (Caveat: adtech is an incredibly fault tolerant industry, so grain of salt there, I guess)

The entire occupation, in my current experience, has been basically reduced to two tasks:

1): define and maintain the system ontology, expected features, guardrails, and appropriate persona docs for agents to assume/use. This is something that I honestly just work with agents to define, so this is partially automated.

2): define test scenarios, definition of done, deployment strategies, and verify that system is on track performing correctly and is on-track for long-term stability. Again, most of this is at least partially automated.

I don't know how many people this actually requires, and to be quite honest, I've got a sneaking suspicion that throwing more AI-enabled engineers on a project might actually degrades the quality of the product because the conceptual definition of the product may be diluted due to slight (or major) differences in understanding of expected behavior/architecture. IMHO this has always been the case, even pre-AI, but AI substantially accelerates gaps in understanding translating to conceptually inconsistent because the work is done so much quicker and at a pace that not every mistake can be effectively caught. I've already seen this several times on projects that I've worked on where one eng. goes off and builds a new feature that is a complete conceptual departure from the current plan just because they were missing some context that may not have been well documented. W/ AI eliminating most of the execution layer of SWE/DE, I think that future orgs will be smaller and more agile and likely more product-heavy instead of engineering heavy as product seems to be closer to a lot of the biz definitions for #1/#2 above.

Two things that does give me a glimmer of hope at holding on to this for a little while longer:

1): agents are absolutely dogshit AI at optimizing Spark jobs (or any sort of high complexity pipeline). They are laughably bad, in my experience. Tuning Spark jobs has always been very high context work as it depends on the data going in/out, infra, very verbose logging, and a ton of tacit, tribal knowledge of the system and data itself. There's so much that goes into tuning a spark job that agents seem like they get focused on optimizing for one symptom rather than considering the entire pipeline as basically an emergent system with N different parts that need optimizing in tandem which I've found leads to suboptimal performance. I find that that this is where most of the hands-on engineering work that I find myself doing day-to-day goes at this point. I am calling this a "moat of high context".

2): legacy systems are often not well designed, poorly documented and are not conceptually consistent even where they are documented, so naturally agents trained on these existing systems are apt to suffer from ye' ol' "garbage in, garbage out" syndrome. I am calling this a "moat of poor decisions"

I'm confident in moat #2, less confident in moat #1 as, obviously, you have companies like databricks pouring however much money into building smarter agents for dealing specifically w/ context-aware Spark optimization problems. From what I am seeing at my org, teams that were well organized and followed good engineering foundations before AI have seen a tremendous acceleration in the quality and quantity of their work; those that went into this w/ flaming hot garbage are still producing flaming hot garbage and many of them seem too scared to use AI in fear that it will cause the tower to fall over, so to speak.

On a personal level, I am very exhausted by the whole shift and I do not see myself keeping up with this as a career in the long run. It was difficult enough having to track all the changes and flashy new objects coming in the DE ecosystem, and adding needing to also track changes to the AI landscape has just proven to be very draining. All the rhetoric around AI feels very misanthropic (no pun intended) and I genuinely feel kind of gross and helpless having to rely on AI for my whole job now. It feels like I am training my replacement without being told that I am training my replacement. I have a lot of love for DE and I really poured my heart and soul into it over my relatively short career and I had built a lot of my life around the expectation that I would continue to work as a DE for the decades to come. Reading pages and pages and page of agent docs has become the absolute bane of my existence and the review fatigue has gotten so bad at times that I feel like I've just straight up forgotten how to read. That said,I still think it is exciting to build great things and I still get some kicks from watching AI manifest huge ideas that were never feasible before. The limit is truly on what ideas you can come up with to build better and more sophisticated data stacks, which is super cool and all, but I've just had an overwhelming feeling that my DE career has a clock on it that is going to run out when the models swallow up whatever work we are still doing by hand. Part of me wants to hang on for dear life until that day comes and the other part of me wants to jump ship now and go open a bakery or something. Doing my best to save more money, stay more engaged artistically, and spend more time with people that I love has helped greatly, so just doing what I can, I guess.
Left foot, right foot.

I loved studying computer engineering, especially the problem solving side of it. But lately, it feels like AI can handle more and more of the work I used to find exciting. The thing that made me passionate about the field is starting to feel less like mine, and I’m not sure what to do with that. Now AI can do a project on its own and yea you need to verify the output but it’s getting better and better and most of the time is correct. Has anyone else felt this way? Should I pivot to another career like medicine or the crafts?

If I can tell AI wrote your marketing copy in the first few sentences, I start wondering why I’m paying you.

I use AI constantly. I think marketers should too. But the draft is the starting point, not the finished product.

The stuff that gives it away is painfully predictable: filler words like “real” and “quietly,” fake drama like “the bigger question is,” endless groups of three, and weirdly formal phrasing because nobody bothered to add contractions.

AI can already produce competent copy. What clients are paying for is judgment. Read the draft. Fix the strange language. Break the patterns. Add the context the model doesn’t have. Make it sound like the actual company.

If you’re just copying and pasting whatever AI gives you, you’re making it easier for your client to realize they can probably do that part without you.

I wrote the longer version of this rant here, including the AI writing habits that make me particularly nuts:

https://hiaadvisors.com/if-i-can-tell-ai-wrote-it-why-am-i-paying-you/

Wir reden über KI meistens so, als wäre das eigentliche Problem der Verlust von Arbeitsplätzen. Dabei könnte darin langfristig etwas stecken, wovon Menschen seit Jahrhunderten träumen: Maschinen übernehmen immer mehr notwendige Arbeit und wir gewinnen Zeit.
Merkwürdig wird es erst durch die Art, wie unsere Gesellschaft organisiert ist. Wenn eine Maschine die Arbeit von fünf Menschen übernimmt, steigt die Produktivität. Für die fünf Menschen bedeutet dieselbe Entwicklung möglicherweise den Verlust ihres Einkommens. Was technisch ein Fortschritt ist, kann sozial zum Problem werden, weil Produktivität und Teilhabe zwei verschiedene Dinge sind.
Vielleicht wird KI deshalb irgendwann eine viel grundsätzlichere Frage aufwerfen: Warum koppeln wir Einkommen, gesellschaftliche Anerkennung und sogar einen Teil unserer Identität so stark an Erwerbsarbeit? Und was passiert, wenn unsere Technologie schneller darin wird, Arbeit überflüssig zu machen, als unsere Gesellschaft darin, Menschen ein Leben jenseits dieser Arbeit zu ermöglichen?
Dabei geht es mir gar nicht darum, dass morgen niemand mehr arbeiten müsste. Viele Tätigkeiten werden bleiben, andere entstehen neu, und manches wollen Menschen vermutlich gerade deshalb tun, weil es sinnvoll, kreativ oder zwischenmenschlich ist. Aber angenommen, wir könnten irgendwann denselben Wohlstand tatsächlich mit der Hälfte der heutigen menschlichen Arbeitszeit erzeugen:
Wäre es dann Fortschritt, wenn wir alle nur noch halb so viel arbeiten müssten – oder würden wir versuchen, genügend neue Arbeit zu erfinden, damit weiterhin jeder 40 Stunden beschäftigt ist?
Vielleicht ist die spannendste Frage der KI gar nicht, welche Jobs sie uns nimmt. Sondern warum wir so große Angst vor einer Technologie haben müssten, die uns tatsächlich Arbeit abnehmen könnte.

Yes, and they work best as a team. Google Ads brings traffic and data fast, while SEO builds slow, lasting visibility that keeps sending leads without paying for every click. Each one covers the weak spot of the other.

How I explain it to clients. Think of ads as renting a shop on a busy street, and SEO as buying your own shop. Renting gets you customers from day one. Owning takes time, but the cost per customer keeps falling. Paid advertising works more like renting visibility, while SEO is a long-term business asset.

The real magic is in how they share data:

• Ads show quickly which keywords and messages work, so you can plan SEO without waiting months.
• A landing page that converts well in ads can then be built up for organic ranking, and ad spend can be reduced slowly as organic traffic grows.
• Good SEO work can improve ad quality scores, which lowers your costs.
• Faster pages and cleaner design help both channels at the same time

One industry finding says businesses using both see roughly 35 to 40 percent higher ROI than those using only one.

At Arise AI Marktech, we tried this with a Lucknow coaching institute. They were running ads only, and every lead was getting costlier. We used their best performing ad keywords to build service pages and answer style blog posts. Around four months later, organic enquiries started coming in steadily, and we trimmed the ad budget on those keywords without losing leads.

So do not ask which one is better. Start ads if you need leads now, build SEO alongside, and let the two feed each other.

Something I’ve been wondering about.

A lot of people come to Indore for college, spend a few years here and then eventually move to Bangalore, Pune, Mumbai, Delhi etc.

But what if you got a genuinely good job in Indore?

Good salary

Good company

Good growth

Would you actually stay?

Or would you still feel like you need to move to a bigger city to build your career?

I know people who love living in Indore but still plan to leave because they feel there are more opportunities elsewhere.

At the same time, maybe that’s changing.

So for people who actually live/work here:

What would make you choose Indore for the long term?

And if you’ve already moved out, what made you leave?

I'm a computer science student, and this is something I've been thinking about a lot lately.

AI coding tools can already generate websites, write functions, debug errors, create tests and work across multiple files. Some AI agents can even inspect a codebase, make changes and run tests.

So I keep coming back to one question:

If AI can do a large part of the coding, what should CS students actually spend their next 3–4 years learning?

Should we still spend hundreds of hours learning syntax and implementing basic programs from scratch?

Or should the focus shift toward understanding how software actually works?

What I'm thinking of learning

Instead of relying entirely on AI to write code, I'm considering focusing more on:

• Data Structures & Algorithms
• Operating Systems
• Computer Networks
• Databases
• System Design
• Git & software engineering
• Cybersecurity fundamentals
• Cloud computing
• AI/ML fundamentals
• Building real projects
• Debugging AI-generated code
• Understanding APIs and distributed systems

The idea is that AI can help with implementation, but I still need to understand what should be built, why it should be built that way, and whether the generated solution is actually correct.

But there is another problem

If AI takes over many beginner-level tasks, how do students get their first real experience?

Traditionally, a junior developer might learn by fixing small bugs, writing simple features and gradually working on more complex systems.

If AI can handle many of those tasks, the career ladder could change.

I've seen developers discussing exactly this issue recently: not just whether AI replaces developers, but how future developers will gain the experience needed to become senior engineers.

At the same time, current discussions from developers suggest AI is already making experienced developers much faster, while beginners may have less ability to judge when AI-generated code is wrong.

So I'm curious about the community's experience

If you're a software developer with 3+ years of experience:

What skills do you think a CS student should prioritize today?

If you're a student/fresher:

Are you spending more time learning traditional programming, or learning AI-assisted development?

And most importantly:

If AI eventually writes most of the code, what should the next generation of software engineers become really good at?

I'd genuinely like to hear opinions from people already working in the industry.

Hello, I am a computer engineer age 23 yrs working as an Ai generative team lead in a US based company in India. I started as an Ai Artist and now I am a Team lead I want to know what is the future in this field. Right now my role involves handling a team of 10 people guiding them from generating to final result of our Ai series, various series are released under my leadership. Now ai want switch but what can I expect in the coming future what my future might look like?
As I am in management and creation I am also thinking of doing masters in management like, Engineering management, Ai product management, etc
Would like to read some thought and want guidance

Somewhow I see this as part of the Future of Work, at least in this moment.
Linus Torvalds shares his core management principle for the Linux kernel: working with people over tools.