bendtb

Hi HN,

I recently finished writing a book (in English), and I would like to turn it into a free eBook. I am Dutch and not particularly fond of my recorded voice, so I am experimenting with three audio formats—all using AI: 1. My voice cloned with AI: I will record a sample, then let the AI generate the full narration in my voice. 2. Standard AI narration: A clear, neutral voice that reads the book word-for-word like a traditional audiobook. 3. Podcast-style version: Using AI to rewrite and narrate the material in a more conversational and digestible format, like a podcast episode.

Curious to hear if anyone has done something similar, and open to recommendations on tools/platforms for any of the above.

Pleased to hear, thank you.

I work for a commmodity / trading / shipping company. We wanted to invest and bring the whole management group for 5 day trip to Silicon Valley, focused on AI and LLM technologies.

The idea was to expose a cross-functional group to the epicentre of innovation – to learn how leading tech companies actually use AI commercially, and bring back insights we could action.

The price tag? USD 500,000 for a week-long program for 10 people. That included visits, some programming, and facilitation. Unsurprisingly, we decided not to go ahead – but the desire to learn at the source is still very much alive. Our idea of a reasonable price tag is around USD 150,000 (where we recon half will go to flights and hotels).

So I turn to the Hacker News community:

1. Is it possible to arrange meaningful company visits to places like Google, OpenAI, or Anthropic without going through a costly middleman? How accessible are these companies for learning-focused groups?

2. Are top-tier universities like Stanford or Berkeley open to group visits, guest lectures, or curated AI sessions for international teams?

3. Are there local Bay Area companies or freelancers that help organise high-impact, cost-effective tech immersion trips?

4. Have you done this kind of trip before — what worked, what flopped, and what should we absolutely not miss if we self-organise?

Any advice, contacts, or shared experiences are welcome.

Thanks!

I work in a project X style department at a mid sized European company. I have designed and built (with the help of an external software consultancy) a software system that updates global users on market prices within our industry, doing various types of calculations with the input I provide. The software is great, but the input of market prices feeding into the system is too little for the software to really kick off and be the best.

In order to increase the quality of the market prices I therefore need to engage a global broker within my industry to get them to update prices globally, which would make my software not only the best in the world, but build a moat around the idea as no one can make something much better.

There are two problems: 1. What do I offer the partner in return? 1. Answer: equity, money, exposure, etc. I am not so bothered about reaching an agreement with the c suite of the company as the idea is good and relatively risk free for them. Also, management love big data project and “digitalization”. 2. This is a very manual process, so the 50-100-200 employees of my partner would need to daily go into a dashboard and plug in 10 numbers each. Not a huge effort, but I have a hard time seeing what could incentivize them, after all they are paid on comission they make and not inputs. This is where I feel it would fall flat, as no inputs make the software less valuable - the whole idea is to sell “live” prices anywhere in the world.

How would you guys think about such a problem?

Hi HN,

I have around 1000 research papers saved over the years that I would like to be able to search in, in a smarter manner than using CTRL+F. This is a personal tool for me to use doing work, and will not be shared with anyone else. It will run on my PC with 4 ghz cpu, 32 gb ram and 8 gb vram/gpu. Nvidia gpu.

I tried Pautobot but it is somewhat slow, which probably has something to do with my hardware. I also tried GPT4all but I couldn’t quite understand how I could train it on my data.

Basically I am looking for a search feature that can rank and return the best matches/paragraphs from these articles. I think vector searching sounds like the better option, but I have trouble finding an easy solution that involves vectorizing PDFs and then being able to search in them documents.

Pleased to hear your inputs

Note: I cannot write code.

Link: https://github.com/nrl-ai/pautobot/

Hello,

I am responsible for a software project at work where I need to create an software solution where commodity traders globally can daily input rates, they will need to maintain around 700 time series x 12 months. That is a lot, but all data will not be updated every time. There need to be a bit of manipulation in the software as rates need to be computated from A to B in that system, as traders need to visually see the rates that their inputs produces. These rates then get sent to a main SQL server, used for various calculations and then displayed in some simple tables.

Currently we have a solution that runs on Excel, but that is unstable. I am trying to frame the questions related to this project the right way, as I am unsure exactly how to achieve my goal (ie a reliable, robust and easy to use input system).

Do any of you have any good advice - or perhaps suggestions for custom systems, Microsoft solutions or Excel hacks that could help me solve my problem?

Hello,

Why do some markets get digitized with direct business to consumer (eg books, cars and stocks) but others still need a broker in between (eg commodities, housing)?

It cannot be the financial size of the transaction, as banks do a lot without any humans involved. On the other hand, commodity markets are almost exclusively using brokers. Does it have to do with how standard a product it is perhaps? Although commodities should be quite standard…

I tried searching google scholar and scihub, but I couldn’t find anything that gave me any reasonable explanation.

Dear all,

I work at a physical commodity operation where we move around 8-10 million metric times of goods CFR. In this connection we are looking to digitize and automate many things in the company which I will not bore you with. However, we are also looking to step up our SnD analysis by incorporating various type of big data analysis, e.g. satellite data. There are some SaaS providers that offers 25% each of what we are looking for via API, so we need to work with multiple vendors. Therefore I am faced with a choice:

1. Either we buy from four different suppliers and solve our data needs quite fast, or… 2. We buy more raw data and invest in building the algorithms ourselves

There is a data science team of two people with a calendar that is 80-90% full already. Therefore I am trying to review the pro’s and con’s of building myself:

Pro’s: - Full flexibility - I own the algorithms - New knowledge created in the organization which could lead to more innovation - No risk of vendor login

Con’s: - Higher costs - A lot of man hours spent - Unknown time frame for implementation - We need extra people to maintain the algorithms

I estimate that the absolute minimum cost of data for building will cost around USD 100.000 (without extra labor costs) and for buying off the shelf USD 150.000.

I would be happy to hear if anyone in here is struggling with the same hurdles and discuss how they are approaching it?