There is a particular kind of anxiety I have experienced more than once. It occurs when I walk into a busy restaurant looking for a place to sit down and then realise I’ll be sitting there by myself while everyone around me is having dinner with someone else.

I prefer spending the entire weekend alone in my place. I have no plans and don’t want to see anyone else. This makes me feel quite peaceful. During this time, I may do whatever I want — read a book, go through social media on my phone, prepare a modest dinner, or even look at the ceiling. Spending the entire weekend alone does not make me feel lonely. In fact, it seems pretty pleasurable. I often look forward to these kinds of weekends.

What if you go into a crowded café to eat alone? Suddenly, I am aware of my hands. Where should I look when I bite? Should I play with my phone to appear busy? Or will it just make me look more pathetic? Are the two across the room looking at me with pity, or am I making up what they might be thinking?

I pretend to be busy, lazily scrolling through my phone as the noodles I ordered start to cool. And in that awkward stillness, a dilemma arises: am I actually terrified of being alone, or do I fear being seen alone?

Solitude is not a problem

Many of us are perfectly comfortable on our own, free from the judgment of others. We sleep alone. We take showers by ourselves. We think alone while we travel, in waiting rooms, or shortly before we go to sleep. Nobody’s looking. Nobody is passing judgment. We deal with these situations without thinking twice.

Solitude itself is neither good nor bad. Most people, especially introverts, don’t perceive alone time as a sign of inferiority but as a way to rejuvenate. It is not a psychological defect to seek serenity or a location free of society’s expectations. This may be an important part of self-care. So why do we feel differently when we eat lunch alone in a restaurant?

The presence of others changes everything

And when other people come into the scene, being alone may start to seem like a show. Suddenly being alone in a restaurant becomes a statement you never meant to make. You’re not only eating; you’re telling something in your mind: No one would come with me. No one thought of asking me to do it.

It doesn’t really matter if any of this is true or not, at that time. The story is written in the emptiness between you and the empty chair on the other side of the table.

Perhaps that’s the point. We’re not really terrified of being alone. We’re afraid of looking lonely.

There is a big difference between the two. Loneliness is an internal experience— a feeling of lacking genuine connection. The fear of being seen as lonely, in contrast, is a form of social judgement: a conclusion others may reach about your place in the world. We feel the former in our own self, but we fear the latter in the eyes of others. And we will do most everything to escape that judgement.

We learned to fear what being alone might mean

This fear did not strike the day I sat alone with cooling noodles. It started much earlier, in between the school cafeteria and every family function where someone would ask why I didn’t bring anyone. Long before we knew what loneliness felt like, we were taught that to be publicly alone was something to explain. There she is, eating lunch alone. Is she all right? We never tested the assumption and accepted the verdict.

Between those school lunch tables and those family gatherings, we learned an unspoken rule: if you’re worth knowing, someone else already has you. Company became evidence. An empty chair turned into a question. The easiest way to not answer it was to prohibit anyone from seeing it.

And so we take care of the proof. We go to the restaurant with a friend we sort of don’t want to see. We tell the host, “I’m waiting for someone. We eat quickly, or we don’t go out at all, calling it a preference. After some time, the avoidance becomes so commonplace that we forget ever choosing it.

But truly, what are we protecting?

Back to those cooling noodles. I don’t think I was guarding myself against loneliness. I would happily spend my weekends alone. I was protecting a story. The tale of me being wanted. The story of my schedule filling up of its own accord. The story of never having dinner with an empty chair across from me.

“That story is heavier than we want to admit, and social media makes it heavier.” Every shared table, every group shot, every “my people” caption makes a public record of who gets chosen. Without even realizing it, we begin to see other people’s lives as evidence of belonging, while our own starts to feel like a footnote. Eating alone in public seems to be a gap in that record. And part of us panics about the divide.

But a picture is just a picture. It tells you almost nothing about whether someone is genuinely connected to others. Some folks are always eating with someone, and come home feeling invisible. Others have friendships that sustain them and eat alone every Tuesday. The table doesn’t tell much. We guard the image because it is visible to people. Real connection, what it’s intended to stand for, is invisible from across a room.

Learning to be noticed

I don’t have a cure, and I’m suspicious of anyone who claims to have one. What I have is a small experiment that I keep repeating.

I’d like a table for one, please. I face down my phone, or try to. I see the food, I see the room, and I see what really happens. Nothing, mostly. People are too busy having their own chats and worrying about how they look.

The pitying looks I had expected are few and far between. If somebody does glance over, it is for a second and has no verdict. This is what psychologists call the spotlight effect: we overestimate how much other people are observing us. Most of the audience I feared was an audience I’d constructed myself.

One thing I have learned is that eating alone doesn’t have to indicate something deeper. An empty chair is not a story of rejection. It is not a story that no one desired your company. Sometimes a table is just a table, for eating. And nothing else. We think people must see our aloneness as a sign that something is missing in our life. In fact, most people are too busy dealing with their own anxieties and daily worries to give any thought to someone else’s.

Perhaps the liberation starts when we stop caring about how our presence — or lack of company — might look to strangers. There’s no one to impress, nothing that has to be meticulously maintained. And when we eventually release that imagined judgment, we can get back to something much simpler: eating our meal while it is still warm.

Are We Afraid of Being Alone or Being Seen Alone? was originally published in The Writing Table on Medium, where people are continuing the conversation by highlighting and responding to this story.

Those five words pretty much sum up the ongoing debate I’ve been having with my inner self.

I’m 39. I’ve traveled, lived through countless experiences, and done things I never imagined I would. And now, at this particular point in my life, I find myself wondering what I actually want for my future, who I am, and who I’m becoming.

Because, yes, I feel like I’m transforming into a version of myself I’m still getting to know. An adult woman. A mature woman who wants mature things. Although, if I’m being honest, I’m not entirely sure what any of that means without falling into other people’s definitions of maturity.

I have way too many outside references.

I constantly hear about responsibilities, about what is morally appropriate, and about the things I should no longer allow myself to do at a certain age. As if making certain mistakes were a privilege reserved exclusively for women in their twenties. Like getting drunk, falling in love and getting carried away, or even believing there’s still a little innocence left in me.

Apparently, by now, I should know better. I should be wiser, anticipate everything, and recognize all the evil in the world before it even has a chance to surprise me.

I don’t know if I’m going through an identity crisis. But I feel like I’m entering a stage of life where I want to drink a glass of red wine while reading a book, simply because it makes me feel like an adult. And at the same time, I want to buy myself a Disney World handbag just to show the world that I’m not interested in outgrowing myself.

Or maybe I should just keep wearing the colors I love: fuchsia, pink, sequins, and all those pretty little details that make me happy. But then I wonder if I’d look like a teenager trapped in the body of a 39-year-old woman who has already had a child.

I think about the looks, the criticism, and that inevitable “That’s for little girls” I might get from people. Especially the other comment that seems to carry so much weight: “You’re too old for that.”

I’ve started thinking that my age is somehow taking away my right to be genuine. That rebelling against certain expectations might make me look difficult or problematic. And although I keep trying to convince myself that I don’t care what other people think, the truth is, I do. Especially when it comes to how they perceive my maturity.

When did being genuine become a sign of immaturity? When did living according to other people’s opinions, thoughts, and feelings become more important than living according to my own?

Another thing that’s been happening since I started studying Psychology is that something about the way I look at relationships has changed. I don’t know if my eyes have simply opened or if my mind has experienced some kind of mental light-switch moment, but I’ve started to understand something I had never really considered before: what it means to have a healthy romantic relationship and, more importantly, what it means to have an emotionally and psychologically healthy person by your side.

Yes, healthy.

I’m not talking about a man I’m simply attracted to, someone who smells good, dresses well, or has a style I like. Nor am I talking about someone who has a good heart or shares certain interests with me. I’m talking about someone who genuinely takes responsibility for his own psychological well-being; someone who makes an effort to improve his habits, his mental health, and the way he relates to other people.

But above all, someone who is willing to build a happy life without falling into toxic positivity. Someone who strives for a kind of happiness that remains grounded in reality.

That idea completely blew my mind.

I had never considered that I might need something like that in my life. I had never really thought about a man being someone who could bring health and inspiration into my life, beyond his intellectual abilities or intelligence.

And I don’t quite know how to explain the feeling of having met a man who made me realize that there is a different kind of person out there. Someone with a different way of approaching life and relating to the world. Someone so different from everything I’ve known, even from the men I’ve been involved with before, including the man I once considered the love of my life.

I’m beginning to understand that love can actually be healthy. That it isn’t just about experiencing an intense feeling, having butterflies, or completely losing your head over someone. It also has to do with how two people choose to live, grow, care for themselves, and relate to each other and the world around them.

At 39, and in the middle of this transformation I feel taking place within me, what I want is someone I can grow alongside. No idealization, no turning him into my savior, and no expecting him to have all the answers. Just someone who inspires me to grow as a woman and as a thinker, someone with whom I can walk that path.

And if I ever find myself walking alone and he notices I’m drifting a little, I hope he can help me find my direction again. Or if I notice that he’s losing his way, I hope I can do the same for him.

I want us to be able to catch each other, encourage each other, support each other, and offer one another empathy as we navigate our own journeys of resilience.

I don’t want someone who will shield me from life’s difficulties. I want someone with whom we can learn to navigate them in a healthier way.

I think I’ve already met that man…

The man who blew my mind.

And that unsettles me quite a bit.

Because I don’t know if I’d call it love. It’s not exactly passion or that kind of emotional frenzy we’ve so often learned to confuse with falling in love. It’s something I’ve never experienced quite this way before. Something that makes me curious, challenges my own assumptions, and, above all, has made me rethink what I want to find in someone and what I want to build with another person.

Maybe what surprises me most isn’t that I’ve met someone like that, but that I’ve discovered I’m now capable of recognizing and wanting something different.

And I had to write about it.

The metamorphosis at 39 was originally published in The Writing Table on Medium, where people are continuing the conversation by highlighting and responding to this story.

Google announced Gemini 4 Carbon on October 28, 2026. Here’s what it does, how it scores on benchmarks, what it costs, and who can use it today.

If you’ve been scrolling past another “Google just dropped the most powerful AI ever” headline this week, I get it. I did too. But this one is worth slowing down for, mostly because the real story is more interesting than the hype.

On October 28, 2026, Google announced Gemini 4 Carbon. It’s the company’s latest frontier model, and it arrives about two weeks after Gemini 4 Argon. Google calls it its strongest model so far.

What Carbon is actually built for

Carbon is not a general chatbot upgrade aimed at everyday users. Google designed it around three areas: finding software vulnerabilities, testing whether attackers could exploit them, and writing fixes. It also handles long tasks better. Google says its internal teams watched Carbon work through dead ends while tracing subtle bugs across large codebases, then produce patches that address the root cause rather than the symptom.

That’s a meaningful distinction. Plenty of AI tools can suggest a quick fix. Fewer can stay with a messy problem for days.

The numbers

Google reports Carbon scored 88.2% on SEC-bench Pro for vulnerability discovery, 54.8% on ExploitGym for exploitation, and 71.0% on CWE-bench for patching. The company also says Carbon roughly halved the rate of unsupported claims compared with Argon in its internal testing, which is a claim worth watching as independent results come in.

On the Artificial Analysis Intelligence Index, Carbon ranks #1 at a score of 59. That’s the figure behind the “#1” headline, but it’s one leaderboard, not a verdict on every model in existence. Benchmarks measure specific things, and real-world performance depends on your use case.

So can you use it?

Probably not yet. Access starts with a select group of cyber defenders, and Google says it will expand through a phased program for trusted partners. Developers, businesses, and consumers will get access “as soon as possible,” but no date has been given.

Pricing has been announced at $2 per million input tokens and $10 per million output tokens during an introductory period, rising to $4 and $20 afterward. Cached input is priced 95% lower. Those are announced figures, not live ones. As of late October, Google’s pricing page didn’t list Carbon yet.

Why this matters beyond the benchmarks

Here’s the part I keep thinking about. A model that’s very good at finding security flaws is also, by definition, very good at finding ways to break things. That’s the dual-use problem, and it’s why Google is starting with vetted defenders rather than opening the doors wide. Google also says it’s participating in voluntary U.S. government pre-release testing, which signals how seriously the company is treating the risk.

For security teams, this could be a real shift. Work that once took a specialist weeks might compress into hours. For everyone else, it’s a reminder that AI capability is starting to concentrate in specific, high-stakes areas, not just in chat windows.

What I’d watch next

A few things will tell us more than any launch post:

Whether independent testers reproduce Google’s numbers. When the API opens, how the pricing holds up. Whether the cyber defenders program produces actual vulnerability fixes that are publicly documented. And whether Google publishes a full model card with context window and output limits, which it hasn’t yet.

The takeaway

Gemini 4 Carbon is a serious model with a narrow launch, strong security-focused results, and one headline ranking that deserves context. It’s not a tool most of us can open today. But it’s a useful signal of where frontier AI is heading: deeper into specialized, long-running work, and more carefully rolled out.

If you work in security, software, or AI, keep an eye on the next few weeks. If you’re just curious, the lesson is simpler: when a headline says “#1 across all models,” check which test it’s talking about.

What’s your take? Are you more excited about faster security work, or more cautious about who gets access first? Let me know in the comments.

Google Just Announced Gemini 4 Carbon. Here’s What It Means for the Rest of Us. was originally published in The Writing Table on Medium, where people are continuing the conversation by highlighting and responding to this story.

The NG (nasogastric) tube streams from her nose to her Daddy’s hand, fortified milk flowing with gravity to her tiny stomach. Business as usual. Except it isn’t.

Mummy’s just made, for the first time, some actual, real food for Poppy to try! Some tasty, proper REAL food! It’s cottage pie (blended to resemble the consistency of a milkshake, of course), and it took me all afternoon. I’m so excited. I’ve been passionate about food my whole life, a cook forever and a chef for years, and finally, I’m going to share it with my girl. I’m sick with nerves. She was never able to manage breastfeeding, a bottle, or a dummy. Her swallowing is medically unsafe; she is severely dysphagic, with the NG tube the only safe way to feed. What if I hurt her? WHAT IF SHE CHOKES!?

But wait, no, she’s making progress; she’s taking interest, she’s been starting to suckle on things, taking tentative dabs of disgusting baby food. She even tried ketchup the other day. That was OK; surely we can try? Surely we have to? I scoop the tiniest spoonful of puree, on the end of the tiniest spoon, and place it, heart pounding, on the end of her tongue.

She retches, she coughs, she splutters and struggles to catch her breath. Daddy pounds her on the back; she’s fine. I cry all afternoon. The puree is in the bin.

Tonight, she’s having giant fishcakes with swirly potato and sweet potato mash (her favourite) and broccoli. And ketchup, of course.

It took three years. Three long years of hard work, of determination, of cutting up mounds of spaghetti into millimetre-long pieces (my favourite).

But she learned to eat.

It is Saturday night, the 8th of June. I’m home with the kids; Daddy’s at work. Poppy is in bed. She drifted off hours ago. Business as usual, except it isn’t.

I’m frantic. Evie, Poppy’s little sister who came along in spring 2024, just won’t settle.

Usually, this is no bother. I’d been sitting up with her until three, four, sometimes later in the morning for a couple of months now. But tonight, I can’t just pour myself another coffee and calmly sit it out. I’ve got work to do, and it’s sitting like an anvil on my chest.

I pour the coffee anyway. Doesn’t help. I am shaking, and not from the caffeine or lack thereof. I can’t breathe steady. My heart is pounding. The Tommy Tippee Perfect Prep machine buzzes in the background, making a milk I hope to heaven will send her to sleep. I feel sick to my stomach, not even tasting the sips of scalding Nescafé. The squealing baby is not helping. The ticking clock is not helping. My frantic texts to my husband, sporadically replying, elbows deep in work as usual, are not helping.

Four or five?

Five. DO IT.

The longer she whines, the harder this is getting. The feeling of dread building, the numbers going round and round in my head like a pencil sharpener, cutting and cutting closer to the core, things feeling sharper and sharper.

Maybe just four. No, maybe none.

Ten PM. She’s just about settling. Now there really is no excuse.

I pour another coffee. Kill time, send another panicked text. Look up the numbers another hundred times. Eleven PM. She’s definitely asleep now. I could just go to bed. I sip the now-cold coffee and try not to cry. Toy with the wedding ring hanging from a chain around my neck, ever since it fell off my finger for the third time, and I nearly lost it for good.

Four. No, five.

Half eleven. The most terrifying pan of water is boiling away on the side as I throw things in a bowl, frantic, heart pounding, not wanting to look at it, not wanting to be awake. I take a picture, send it, to prove it to myself more than anyone, to make it concrete.

Midnight.

I have five. Five whole prawns.

I’m having eight tonight. And some chicken breast and a large portion of rice. And a load of vegetables. And some pretty awesome teriyaki sauce.

I’ve still crunched the stupid numbers; I’ve still counted and recounted everything; I still feel sick at the thought. I don’t understand it. I don’t know what, or when, or how. What’s right or wrong, or whether it even matters. My body is confused, my brain more so. My hunger signals are beyond comprehension, my metabolism like a bull in a china shop. I’m obsessed and apathetic, excited and terrified, overjoyed and angry, starving, hungry and painfully nauseous. I’m still learning to eat. But, like my incredible daughter, I’ll learn.

And if this too takes three years, of hard work, of determination, of more mounds of spaghetti (or just teriyaki sauce), it’ll surely be just as worth it.

(makes one small jar)

45g Soft light brown sugar

40g Honey

70ml Soy sauce (I use mostly light, with a few hefty dashes of dark for colour)

Thumb-long piece of root ginger, finely grated

2 Garlic cloves, finely grated

Large pinch chilli flakes

2 tsp Cornflour, mixed with a splash of water

2 tsp Rice vinegar or Lime juice

Small drizzle of sesame oil

Mix the sugar, honey, soy, ginger, garlic and chilli flakes with about 100ml water (don’t be too precious about it, I free pour) in a small saucepan and bring to a simmer, stirring to dissolve the sugar. Bubble for five to ten minutes to reduce slightly and infuse the flavours.

Quickly whisk in the cornflour. The sauce will start to thicken immediately. Keep stirring for a few minutes, then stir in the vinegar or lime juice and sesame oil.

Strain through a sieve and allow to cool.

Use immediately as a glaze for pan-fried or baked meat and fish, as a dipping sauce, as a stir-fry sauce (diluted with a bit of water), or store in the fridge for up to two weeks. This sauce is pretty strong stuff, so a little goes a long way.

A heartfelt thank you to everyone who has supported me by buying a coffee through the button at the top of the page, especially Matt and Kim-your generosity was much needed! We also deeply appreciate all of you who have liked, commented, and subscribed to the blog. Your support means the world to us and helps us continue sharing our journey.

We read and appreciate all your comments and words of encouragement. If you’d like, please feel free to share a few words below.

Originally published at https://mingo.life on July 19, 2024.

The Hardest Thing I’ve Ever Had To Do Is Follow My Daughter’s Example was originally published in The Writing Table on Medium, where people are continuing the conversation by highlighting and responding to this story.

I open my laptop, say “Hi Claude,” and enquire about my pending tasks. Add a few MCPs, adjust the harness, and let Claude do my job. Then I play around with a few agents, delegating them to fix my PRs and investigate bugs.

Sounds perfect, yet terrifying.

Not because AI is replacing jobs and creating a layoff catastrophe. I’m more interested in something much less dramatic. AI is changing what it feels like to work.

It has certainly made life easier. But are we entirely okay with the side effects it is bringing to the table? I’m not quite sure.

I miss sitting with a fellow engineer and asking why a particular service behaves the way it does. Those conversations shaped my thinking. There was always something to learn from someone else’s experience, especially when dealing with systems that had accumulated years of engineering decisions.

I now have more conversations with Claude than with some of my actual colleagues. And it’s not just the conversations I’m missing. It’s the experience of actually doing the work.

I no longer feel the same thrill of fixing a bug or developing a new feature when I can simply summon an agent to do it for me. There was something satisfying about finding a problem yourself, forming a theory, blaming the database, caching or the previous developer (sometimes, unfortunately, it was me), and eventually figuring out a fix. The learning was hidden in that process.

Spending hours understanding why something broke, only to realize that the solution was a ridiculously simple change. Not every solution was worth the time spent. But those experiences gradually helped develop an intuition for how systems behave.

I’ve gone from fixing bugs to assigning bugs to agents, who occasionally introduce new bugs. It’s a beautiful ecosystem. Now, I feel like I’m losing control over my own code. Claude and Codex have become my IDE, and I’m just a content writer for them. Your worth is measured by the tokens you’ve burnt.

I sometimes wonder how a junior engineer is supposed to develop that intuition when the very problems that helped us learn are delegated away.

Most engineers develop their understanding of systems through making mistakes, breaking things, fixing them, reviewing someone else’s code and occasionally asking questions that might sound silly.

That experience is difficult to compress into a prompt or transfer through a generated explanation. And this shift isn’t limited to the individual developer. It’s slowly changing how we work as a team, too.

The estimation and refinement meetings are disappearing because AI is, once again, in the picture. I hated long calls. But some meetings served a purpose. They gave us an overall business picture of why we were building an epic in the first place.

They helped us understand requirements, discuss trade-offs, challenge assumptions and, most importantly, learn from each other. The details often emerge through conversations, disagreements and the occasional engineer asking a question that nobody had thought about.

Now, someone randomly throws AI-generated JIRAs onto the board. What used to be precise and concise, after meaningful discussions, has become content-heavy documentation that requires AI again to summarize and explain.

Human-to-human communication has taken a back seat, with token economics running the show. Thank God, at least my one-to-ones haven’t been replaced by AI agents. Yet.

Of course, nobody wants to spend an entire day fixing a bug that AI could resolve in minutes. But if every interesting problem is delegated before an engineer has the opportunity to understand it, we might be creating a different kind of learning gap.

We could end up with engineers who are excellent at asking AI to solve problems but have fewer opportunities to develop the judgment required to evaluate those solutions. And that judgment becomes increasingly important as the systems we build become more complicated.

There’s another thing I wonder about: ownership.

When you write a piece of code, debug it, understand its dependencies and eventually ship it, you develop a certain sense of responsibility towards it. You know why a particular decision was made and what might break if someone changes it later.

Of course, reviewing AI-generated code is still engineering work. But reviewing something you understand and reviewing something you are trying to understand are two very different experiences.

And when an agent produces like 200 files in a few minutes, nobody is particularly excited about reading every single line. Sometimes, the PR review itself starts feeling like another project.

The human-in-the-loop part thankfully exists in the agentic setup, but it gives me the feeling of being a manager rather than a developer. Technically, the role feels dead when all we do is orchestrate agents, review their output and make GPUs tired with our prompts.

Perhaps that’s an exaggeration. But it does make me wonder what happens to the experience of being an engineer when the interesting parts of engineering are increasingly delegated.

This is not an anti-AI rant. I’m pretty much happy to let AI write the boilerplate, do the documentation, automate repetitive tasks and fix things. But when machines start doing the interesting parts, humans may miss the experience of actually doing the work. And that experience matters.

AI ways of working should keep humans involved where humans actually matter. The conversations, mentoring, debates and context should not be lost in the process of achieving faster deliverables.

Engineers should be kept close enough to the code that they still understand what they are building. The learning is hidden in fixing broken systems, not just in using better prompting techniques.

We are building smarter machines every day. Perhaps it’s equally important to ensure that, in the process, we don’t make the humans behind them a little less curious, creative and connected.

Are We Outsourcing the Joy of Engineering? was originally published in The Writing Table on Medium, where people are continuing the conversation by highlighting and responding to this story.