This isn't something you'd be able to manage with a normal UI since every agent has its own at of settings. My agents come in sets, I usually change a set or multiple sets, 10-15 settings per agent. And my harness hasn't made a single mistake once I worked out all the kinks from MCP.

This is nuts!

First, that chart is super zoomed in. You can't zoom out or switch timeframes. You're not told how many bars you need to predict. You're not told the dates or what the stock is.

Not much you can see there other than a handful of candles, that's doesn't really tell you much.

If the stock and dates are a secret so people don't cheat, I don't see why. People can cheat on Wordle or Worldle, and they're still fun to play. But here you're basically just guessing, you have so little information it's basically a coin flip.

Then, by the time GPT 5.5 came out you could already generate a flawless PDF and Word doc resume with the same formatting as your base resume. It really is amazing.

Trying to compete with these labs is a bad idea, you might work hard on your thin wrapper and then they come out and have their chatbot or harness do it way better than you.

But I also think this is really cool. It's a new era in software. Build something worthwhile and you won't need to worry about a lab doing the same thing with their chatbot.

I've been enjoying the resets, plus I had 3 resets I haven't used.

Just using 5.6 Sol in Fast mode the whole time, 1B tokens per day.

However, when they removed the 5h limit they also quietly lowered the 5.6 Sol context from 354k to 258k or something like that. I noticed it in Codex.

Other than that, I think the difference between 5.5 and 5.6 will be the same as 5.4 and 5.5. 5.5 is just less frustrating to use, although not perfect and still has derp moments. But a lot less than 5.4.

So I expect 5.6 Sol to be smoother to use. But so far it just feels slower. We'll see.

I have tested two judge models in my apps:

1. Judge model for a resume tailor. It evaluated the result resume vs the base resume and JD and judged it out of 10 on fit and honesty. It worked well and was useful.

2. Review model in my LLM trading bot platform. It reviews decisions from the Main model. The problem here is that the bot is navigating ambiguity. So unless the Review model catches an outright blunder (e.g. making a decision on wrong candle price or a BUY when it should be a SELL), the Review model can do more harm than good.

First, it adds latency to decisions, decisions take twice the amount of time (like be 60s instead of 30s for Gemma 4 31B). Second, it can make the bot too cautious, because Review model only runs on BUY/SELL decisions and not HOLD decisions, so the bot will only make less trades instead of review model increasing number of trades (because of latency and cost).

So overall, I think you'll get better results with a better model single shotting it rather than a review model if the answer isn't easily verifiable. But then why do you need a judge model and not just have the same agent review itself?

ALSO, if you read the reasoning text for a reasoning model (like Gemma 4), you see that it ALREADY reviews itself. So it's doing its best, re-review isn't really adding information. It's an interesting experiment, but you need to evaluate on a case by case basis.

I think this will flop. Even amazing halo car EVs have poor resale value, and this one isn't it. It will not keep value like an analog Ferrari, but may be better than Rimac because it's a Ferrari and if they limit supply.

I'm all for EVs by the way, I drive a Model 3 Performance and I love it. Just not feeling this design at all.

It was mismanagement then and it's mismanagement now, the more things change the more they stay the same.

You can use it for free, forever, if you just run the bot in your browser (client mode). Server mode is premium, but you don't need it to run the bots.

I posted about it in this comment: https://news.ycombinator.com/item?id=48085993#48088468

We've been playing with different settings and models since January, older wallets have had some bleed with Gemini 3 Flash and bot settings that didn't perform well.

We've been running bots exclusively on Gemma 4 31B and 26B for the last 7-10 days and they've been either breaking even or trading very well, it really depends on the coin. I think around 10-30% gain for the good cases for the week.

It's still experimental, I only put $100 into each bot (and that's what I recommend to my users) so it's not crazy money, but once we're comfortable we'll put more money into it.

https://vtxmacro.com/leaderboard?t=24h

How many paying clients (companies, not people) do you have?

The bot settings (system prompt and user prompt, temperature, reasoning, etc.) are 100% transparent and customizable, and all users can view and copy anyone else's settings from the leaderboard. The goal is to build the best trading bots possible by seeing what works.

You can run a bot on Gemini 4 31B with a free tier Google AI Studio account (I'm running 5 bots on it myself). Or just run Gemma 4 26B on your PC if you have the GPU for it. I'm running 5 on my 5090, so I'm trading with 10 bots total.

The platform is connected to Hyperliquid and you can trace all the trades on the blockchain from the user's Analytics page (always public).

The way it works is you set a loop interval (default 1 minute) and the model receives the candles, market stats, indicators, account balance, current positions and so on and decides Buy, Sell, or Hold and how many units.

It's still experimental but I have already processed 1m+ prompts, 10k+ trades, and almost $1m in volume since January 2026. I have around 15 bots running right now, you can check their PnL on the leaderboard (public). I've made a lot of changes in the last few weeks so most recent either 24h or 7d results are the most relevant. The model you use is super important (Gemma 4 31B so far is the best value I found, better than Gemini 3 Flash and you can run it for free) and also the coin you choose is important too. Preferably, you want something that's trending. My friend's bot did well with ZEC and VVV this week.

Right now I'm working on improving reliability (I bought a Japanese VPS to run my own HL node), and this weekend I moved the app from Render to my own DC VPS for 10x+ cheaper and 1000x more bandwidth (25 TB instead of 25 GB, seriously if you're using Render and want cheaper infra look into buying your own VPS).

I'm also implementing CLI/MCP for OpenClaw support. And next is an automatic screener that will use LLMs to pick the most promising cryptos to trade (since I noticed this has a huge effect on PnL).

If you have questions, let me know, the Trade page has my Telegram group link.

I'm using Gemma 4 31B in my app with 5 agents, 1.5k requests per day, each.

I was using 100M+ tokens per day, $250 per day or so and only paying $160 per month to GitHub.

I cancelled my GHCP sub and switched to Codex last week, so far so good but I miss Gemini 3.1 Pro for UI work.

Also, the premise that it took each of them a year to do the project means Bob was slacking because he probably could've done it in less than a month.