Building long horizon agents@Intuit. entrepreneur & builder.
SFO & BLR
Joined December 2008
- Tweets3K
- Following2K
- Followers676
- Likes2.7K
Inder Singh retweeted
The thing I’m most grateful to Elon for is bringing back an age of wonder and excitement to humanity. That was jaw-dropping awesome! Screams and mouths wide open in the audience. Nuts.
Inder Singh retweeted
Specialized sub-agents seem to be the next emergent pattern for AI software development tools.
Last week, Anthropic published a blog post on how they’re building long-running agents for software development for complex, broad-scoped tasks.
One of the key optimizations they talk about is reducing content in a single session’s context window, which considerably seems to improve agent recall on the task at hand.
anthropic.com/engineering/ef…
Inder Singh retweeted
Nothing good has ever been invented by committee.
“Most inventors and engineers I’ve met are like me—they’re shy and they live in their heads.
They’re almost like artists. In fact, the very best of them are artists. And artists work best alone—best outside of corporate environments, best where they can control an invention’s design without a lot of other people designing it for marketing or some other committee. I don’t believe anything really revolutionary has ever been invented by committee. Because the committee would never agree on it!”
Steve Wozniak is the engineer who quietly built Apple.
Here are 26 ideas I took away from this episode and my research you can use.
1. Constraints force deep understanding.
2. Focus on the step, not the outcome.
3. Committees kill revolutions.
4. Learning is the prize.
5. Institutions by default reject anything that means existing beliefs are wrong.
6. Happiness equals smiles minus frowns.
7. Misplaced loyalty is a waste.
8. Work alone on what matters if you must.
9. Patience compounds.
10. Hold your ideas with the right grip. Let go of incorrect ideas.
11. If it's worth doing, it's worth giving it 100%.
12. Obsession isn't a problem. It's an advantage.
13. Simplicity has the fewest moving parts.
14. Time will do the work for you if you align with how the world works.
15. Move with urgency. You can do it much faster than you think.
16. Design around engineering, not marketing.
17. Optimize for happiness, not fairness
18. You don't have to run the company to be a co-founder.
19. "It takes a lot of work to make something simple."
20. Obsess over customers.
21. Don't accept something because it's the way it is.
22. You win in the dark, when everyone else is partying or sleeping.
23. The only way to understand is to get your hands dirty in the work.
24. "Simplicity is the ultimate sophistication."
25. The best are always learning more.
26. Never lie. Honesty is the most important thing.
(Listen now "Steve Wozniak on The Knowledge Project" or see links in comments.)
Inder Singh retweeted
“The disease of our times is that we live on the surface. We’re like the Platte River, a mile wide and an inch deep. I always say, 'If you want to become a billionaire, invent something that will allow people to indulge their own Resistance.' Somebody did invent it. It’s called the Internet. Social media. That wonderland where we can flit from one superficial, jerkoff distraction to another, always remaining on the surface, never going deeper than an inch. Real work and real satisfaction come from the opposite of what the web provides. They come from going deep into something—the book you’re writing, the album, the movie—and staying there for a long, long time.”
— Steven Pressfield in Tribe of Mentors
Inder Singh retweeted
Elon Musk's formula for success: Ego ÷ Ability < 1
When your ego exceeds your ability, you break the feedback loop to reality. Smash your ego. Internalize responsibility. Do whatever it takes.
The goal isn't glory—it's true work that serves the most people.
Inder Singh retweeted
Innovation is art - it can’t be scaled.
Scaling a team creates a hierarchy to reduce coordination costs.
Hierarchy introduces the Principal-Agent problem.
Agents optimize for status and cohesion over correctness.
Going from zero to one requires a founder-led flat team.
Inder Singh retweeted
Be generous in your judgement of others, but ruthless in your judgement of yourself.
Inder Singh retweeted
We have to take the LLMs to school.
When you open any textbook, you'll see three major types of information:
1. Background information / exposition. The meat of the textbook that explains concepts. As you attend over it, your brain is training on that data. This is equivalent to pretraining, where the model is reading the internet and accumulating background knowledge.
2. Worked problems with solutions. These are concrete examples of how an expert solves problems. They are demonstrations to be imitated. This is equivalent to supervised finetuning, where the model is finetuning on "ideal responses" for an Assistant, written by humans.
3. Practice problems. These are prompts to the student, usually without the solution, but always with the final answer. There are usually many, many of these at the end of each chapter. They are prompting the student to learn by trial & error - they have to try a bunch of stuff to get to the right answer. This is equivalent to reinforcement learning.
We've subjected LLMs to a ton of 1 and 2, but 3 is a nascent, emerging frontier. When we're creating datasets for LLMs, it's no different from writing textbooks for them, with these 3 types of data. They have to read, and they have to practice.
Inder Singh retweeted
Behind every great product there’s at least one engineer who knows the entire stack, takes problems personally, and jumps in and solves them.
Without regard to ownership, credit, or chain of command.
If you don’t know who this is in your company, I have bad news for you…
Inder Singh retweeted
“The greatest people are self-managing. They don’t need to be managed. Once they know what to do, they’ll go figure out how to do it. What they need is a common vision, and that’s what leadership is.”
— Steve Jobs
