With a little bit of experience, I realized that it makes sense even for agent to run commands/scripts for deterministic tasks. For example, to find a particular app out of a list of N (can be 100) with a complex filtering crietria, best option is to run a shell command to get specific output.
Like this, you can divide a job to be done into blocks of reasoning and deterministic tasks. The later are scripts/commands. The whole package is called skills.
"I don't trust LLMs to do the kind of precise deterministic work" => I think LLM is not doing the precise arithmetic. It is the agent with lots of knowledge (skills) and tools. Precise deterministic work is done by tools (deterministic code). Skills brings domain knowledge and how to sequence a task. Agent executes it. LLM predicts the next token.
I find it difficult to comprehend if the billionaires do not know this. That if the Government slowly turns into an oligarchy with their help, the Government will eventually turns against them too. They cannot escape the punishment.
Humans (our leaders) have behavior that shows up when under stress or things do not go per their plan. Other countries are already noticing this about USA leaders and will not hesitate to exploit.
My prediction for the future, with a heavy heart.
- Feb/2026: China becomes #1 in technological progress. And America's relationship with former allies is irreversibly bad.
- 2027-29: Russia continues to pamper President Trump for another term. Their goal is to create anarchy in America. They succeed.
The amount of incompetence, incomprehensible ideology and red-tape slows everything down. NSA and CIA still look powerful because of huge amount of funds that flow in. Other countries happily hack into our infrastructure. You hardly hear NSA doing such accomplishments anymore (last was the Iranian nuclear hack). I doubt if America has any powerful spy network in places in Russia or China. But hey, I am just a dumb citizen.
Maybe yes. If so, that is great. But then why do we not see any damage to those countries either due to misinformation or any other hacks. On a related note, Snowden incident shows the depth of our ability.
This is the point. Right tool for the job. Kubernetes was incubated at Google and designed for deployments at scale. Lot of teams are happily using it. But it is definitely not for startups or solo devs, unless you are an expert user already.
Thanks! Helium only automates browsers. If the 2FA is happening in the browser, then you can use Helium to automate the flow. If it's outside, then that part cannot be handled by Helium.
I wonder who came up with the $200/month idea, and what was running in their mind.
$200/month = $2400/year
We (consumers/enterprises) are already accustomed to a baseline price. Their model quality will be caught up or exceeded by open-source in ~6 months. So if I find it difficult to justify paying $20/month, why should I even think about $200/month.
Probably the thought process was that we can package all the great things (text, voice, video, images) and experience. The problem is that very few people use everything. Most of the time, the use cases are limited. Someone wants to use for coding, while someone else (artist) wants to use for Sora. OpenAI had an opportunity to introduce a la carte pricing, and then go to bundling. My hypothesis is that they will have very few takers at $200 for the bundle.
Enterprises - did they interview enterprises enough to see if they need user licenses for the bundles? Maybe they will give it at 80% or 90% discount to drive adoption.
Disclosure:
I am on Claude, Grok 2/X Pro, Cursor Personal, and Github Copilot enterprise. My ChatGPT monthly subscription expires in a week, and I will not renew for now and see the user vibes before deciding. I have limited brain power to multitask between so many models, and I will rather give a chance to Gemini Pro for 6 months.
1/ Best-in-class LLM in Bedrock. This could be done w/o the partnership as well.
2/ Evolving Tranium and Inferential as worthy competitors for large scale training and inference. They have thousands of large-scale customers, and as the adoption grows, the investment will pay for itself.
reply