General
p/generalShare and discuss tech, products, business, startups, or product recommendations
trending
•

1yr ago

3 Months of Hard Work, Zero Retention

Hey everyone, hope you're doing well this weekend.

I m not here to promote anything. Just feeling a bit stuck and wanted to share something.

I recently built an MVP: a productivity tool I originally made for myself to stay focused.
I really thought others might find it useful too. A few people tried it and gave positive feedback, but no one really stuck around to use it.

It s a weird place to be. I ve spent months working on it, sacrificing all my free time after my full-time job, and now I m reconsidering whether to keep going or just trash the whole thing (honestly leaning more toward the second ).

•

2yr ago

Is Listening the Secret to Great Communication?

People often say that effective communication is about how well you speak your choice of words, tone, and delivery. But isn t listening just as, if not more, important? Active listening goes beyond just hearing words; it s about truly understanding the emotions, intentions, and ideas behind them. Think about the best conversations you ve had weren t they the ones where the other person really got you? When you re a good listener, you re able to build deeper connections, respond thoughtfully, and even improve how you express yourself. Do you think listening is an underrated superpower in communication? How do you actively practice it? Maybe it s paraphrasing what you ve heard, asking follow-up questions, or simply resisting the urge to interrupt. Let s talk about how listening shapes great communicators and fosters meaningful interactions!"
•

3mo ago

We were optimizing output tokens to save money. Turns out 95% of our bill was input.

Analyzed a day of token logs across an autonomous coding agent setup running on internal codebases. The raw count: 769M input tokens vs 7.4M output tokens (~104:1).

Because long agent runs re-read session history (files, AST diffs, test outputs) every turn, input costs accounted for ~95% of total spend. Optimizing output length turns out to be looking at the wrong variable.

Three things actually saved us money:

  1. Routing: Shifted non-interactive workloads (evals, background analysis) to batch/flex channels. Billed at 0.5x list price with zero code logic changes.

  2. Cache Discipline: Kept system prompts strictly byte-stable (no top-level timestamps). Achieved a 94.9% prompt cache share, driving input costs from $10/M down to ~$1.46/M blended.

  3. Context Compression: Built a pipeline sending compact session representations instead of verbatim transcripts. Achieved 2.83x median compression (fitting ~500K session history into a 200K window).

•

3mo ago

Start with a default, then expose the setting

Many products add a setting as soon as two users ask for different behaviour. The result is often a page of switches that makes the common task harder. A better first step is to identify the most frequent, least surprising default and leave the edge case visible in the decision record. When have you chosen a strong default instead of an early setting, and what told you it was safe to do so?

Hey Emoji lovers, do you know about Unicode? 🧡

Hey fellow PH-ers, I see that almost everyone is using emojis for almost every purpose. Do you know that the organization that is standardizing them and keeping them alive is called Unicode and you can help support them by "adopting" your favorite emoji? https://aac.unicode.org/ I already adopted my favorite one :)
•

4mo ago

If you run AI agents in production, how do they actually get access to your systems?

I have been thinking a lot about how teams give their AI agents access to real systems, and I would rather hear how people are doing it in practice than in theory. On one end I see teams hand an agent a single API key or service credential that can basically touch everything, because it was the fastest way to ship. On the other end I see people trying to give each agent its own scoped identity, short-lived tokens, and some kind of audit trail, which is a lot more work to stand up. Most teams seem to sit somewhere uncomfortable in the middle. If you are running agents in production today, what does your setup actually look like, how did you land on it, and is there a part of it that quietly worries you when you think about it late at night?

•

2yr ago

AWS SageMaker vs. Amazon Machine Learning

Looking to understand the differences between AWS's machine learning offerings? I've written a comprehensive comparison between AWS SageMaker and Amazon ML. This article explores: SageMaker's end-to-end ML capabilities vs. Amazon ML's historical approach Detailed analysis of SageMaker's features including Studio, Autopilot, and Ground Truth Implementation best practices and cost optimization strategies Practical recommendations for teams starting with machine learning on AWS Whether you're a data scientist, developer, or decision-maker, this comparison will help you understand AWS's machine learning evolution and make informed decisions about using SageMaker for your ML projects. Here is the article link - https://medium.com/@smith.emily2... Join the discussion and share your experiences with these platforms!
•

4mo ago

How much user feedback is enough before launching a new feature?

Hi everyone,

I'm curious how other founders decide when a feature is ready for release.

Do you rely on feedback from a small group of beta users, or do you wait until you have a larger sample size before making the feature public?

I've seen cases where launching early helps gather valuable insights, but I've also seen users become frustrated when a feature still has rough edges.

First
Previous
•••
527528529
•••
Next