Nikita Kurpas

Software Engineer @IndeedJapan, freelance @toptal

[James Clear @JamesClear]

The recipe for frustration is taking on something that requires a whole person, but only giving it half your time.

Look at what you're trying to do. What is required? Some things can only be done well when they are pursued with full effort.

[Nikita Kurpas @nikkurpas]

@thsottiaux I’m on a Plus plan and I found it way better than having 5h limits. I can actually have a few intense coding sessions per week now. The 5h limits before were too restrictive and ran out pretty quickly. With that gone, I found myself doing more work with Codex when I have the time

[OpenAI Developers @OpenAIDevs]

https://t.co/6Y1KwxpJjj

[Artificial Analysis @ArtificialAnlys]

GPT-5.6 Sol and Luna are ahead of Terra at every point on the Intelligence vs Cost per Task chart. GPT-5.6 Luna stands out as a particularly cost efficient model

Charting the Artificial Analysis Intelligence Index shows the trade-off between intelligence and Cost per Intelligence Index Task. Across reasoning efforts, each GPT-5.6 model pushes past GPT-5.5 on the Pareto frontier (excluding non-reasoning).

However, Luna and Sol are always ahead of Terra. This means for any Terra effort level, there is a Luna or Sol effort level that is more intelligent at no extra cost, or as intelligent at lower cost.

[James Clear @JamesClear]

In my early 20s, I wrote a list of ambitious things I wanted to do in life. Go bungee jumping. Create a movie. Start a business. That sort of thing. I did a few of them, but not most.

Looking back, however, that list was incredibly valuable because it taught me the usefulness of chasing bold ideas—even if they never materialize. The dreams that didn't happen changed shape and led me down other interesting paths.

The important thing isn't to achieve all your dreams, but to keep dreaming. Your desires change as you learn and grow. Old dreams will spawn new dreams. And eventually, one of these newer, more refined visions will find its moment and become reality.

[James Clear @JamesClear]

In theory, consistency is about being disciplined, determined, and unwavering.

In practice, consistency is about being adaptable. Don't have much time? Scale it down. Don't have much energy? Do the easy version. Find different ways to show up depending on the circumstances. Let your habits change shape to meet the demands of the day.

Adaptability is the way of consistency.

[James Clear @JamesClear]

Be excited for people when they succeed.

When a friend or family member reaches an important milestone like getting a promotion or making their first sale or scoring acceptance into their desired program, celebrate it. Buy them a drink. Send them a card. Tell them you’re proud to know them.

Being thrilled on someone’s behalf is a lovely way to be. Winning is better when shared.

[James Clear @JamesClear]

What is the real goal?

The real goal is not to “beat the market.” The goal is to build wealth.

The real goal is not to read more books. The goal is to understand what you read.

Don’t let a proxy become the target. Don’t optimize for the wrong outcome.

[Mark Manson @Markmanson]

Maturity is the ability to prioritize values before feelings. The child does what feels right, the adult does what is right.

[Superdaddy @jakemor]

🧵 Superwall had on off-the-record* meeting with someone very high up at Apple yesterday to ask questions about what is and isn't allowed on the App Store. We asked for rules on

→ App2Web
→ Transaction Abandon
→ Free Trial Toggles
→ A/B Testing Paywalls in General

Here's everything we learned...

* Guidelines are intentionally subjective and ultimately up to the reviewer to interpret (more on this in my last post). The only way Apple takes an official stance on a guideline is by updating them.

Some topics may never enter the official guidelines... that doesn't mean they're allowed.

There's a whole layer of internal policy behind the guidelines that isn't written down anywhere and subject to change.

The text you read is just the tip of the iceberg. That's what this thread is about. 👇

[Mark Manson @Markmanson]

You are not unhappy because your life is bad. You are unhappy because you have irrational expectations for it to be better.

[James Clear @JamesClear]

Life rarely changes in a positive way without an increase in responsibility.

That can mean taking ownership of your health or committing to a relationship or starting a business.

Whatever it is, if you want the trajectory to change, the amount of responsibility usually has to change.

[Bo Wang @BoWang87]

A complete cell image by chatGPT Image 2.0 🔥

[James Clear @JamesClear]

There are at least 4 types of wealth:

- Financial wealth (money)

- Social wealth (status)

- Time wealth (freedom)

- Physical wealth (health)

Be wary of jobs that lure you in with 1 and 2, but rob you of 3 and 4.

[James Clear @JamesClear]

So much of what we believe is the residue of someone else's thinking.

Pause and question things for a moment. Is this really how it has to be? Is this really what you want?

[Flavio Adamo @flavioAd]

Ever wondered how much that feature actually cost you to build with Codex?

No? Me neither. But I built an app to track it anyway

this app ended up costing me €9.77 btw

[James Clear @JamesClear]

It only takes five minutes to break the cycle.

Five minutes of exercise and you are back on the path. Five minutes of writing and the manuscript is moving forward again. Five minutes of conversation and the relationship is restored.

It doesn't take much to feel good again.

[vogel @ryanvogel]

🚨 AGENT WORKFLOW UPDATE 🚨

Plan Mode: GPT-5.4
Build Mode: GPT 5.3 Codex
Subagent (Explore/Docs/Second Opinion): GPT 5.3 Codex Spark

This is the first time I have no Claude based models in my workflow. I feel a lot more productive too (while spending way less)

[Andrej Karpathy @karpathy]

It is hard to communicate how much programming has changed due to AI in the last 2 months: not gradually and over time in the "progress as usual" way, but specifically this last December. There are a number of asterisks but imo coding agents basically didn’t work before December and basically work since - the models have significantly higher quality, long-term coherence and tenacity and they can power through large and long tasks, well past enough that it is extremely disruptive to the default programming workflow.

Just to give an example, over the weekend I was building a local video analysis dashboard for the cameras of my home so I wrote: “Here is the local IP and username/password of my DGX Spark. Log in, set up ssh keys, set up vLLM, download and bench Qwen3-VL, set up a server endpoint to inference videos, a basic web ui dashboard, test everything, set it up with systemd, record memory notes for yourself and write up a markdown report for me”. The agent went off for ~30 minutes, ran into multiple issues, researched solutions online, resolved them one by one, wrote the code, tested it, debugged it, set up the services, and came back with the report and it was just done. I didn’t touch anything. All of this could easily have been a weekend project just 3 months ago but today it’s something you kick off and forget about for 30 minutes.

As a result, programming is becoming unrecognizable. You’re not typing computer code into an editor like the way things were since computers were invented, that era is over. You're spinning up AI agents, giving them tasks *in English* and managing and reviewing their work in parallel. The biggest prize is in figuring out how you can keep ascending the layers of abstraction to set up long-running orchestrator Claws with all of the right tools, memory and instructions that productively manage multiple parallel Code instances for you. The leverage achievable via top tier "agentic engineering" feels very high right now.

It’s not perfect, it needs high-level direction, judgement, taste, oversight, iteration and hints and ideas. It works a lot better in some scenarios than others (e.g. especially for tasks that are well-specified and where you can verify/test functionality). The key is to build intuition to decompose the task just right to hand off the parts that work and help out around the edges. But imo, this is nowhere near "business as usual" time in software.

[James Clear @JamesClear]

Environment can multiply (or divide) your effort.

Before you plant the garden, study the soil. After you plant the garden, tend the weeds.

First, selection. Think carefully about the right person to marry or business to pursue or city to live in.

Then, cultivation. Once you choose, care for it lovingly and nurture it into the best it can be.