Polymath Engineer Weekly #75
Transformers, Rust allocations, Prompt Engineering, Debezium, OCaml, Techno-feudalism and Occam's Razor
Links of the week
In this post, we will look at The Transformer – a model that uses attention to boost the speed with which these models can be trained. The Transformer outperforms the Google Neural Machine Translation model in specific tasks. The biggest benefit, however, comes from how The Transformer lends itself to parallelization.
What does “faster” even mean?
In this case we’re talking about cloning performance, aka how long it takes to call the .clone() function on the element.
In the case of String this means mallocing a new string and then memcopying the data over. Arc<str> by comparison simply increments the strong reference counter.
Prompt engineering is a facet of development with LLMs that has garnered massive amounts of hype throughout the industry. Social media is flooded with AI influencers claiming to have created the best prompts for x or y tasks, new guides pop up every day on how to effectively guide LLM output, and in July, Netflix listed a Prompt Engineer position with a staggering salary of $900,000 per year. Despite all of the buzz, prompt engineering is a relatively straightforward but incredibly important concept in harnessing the power of large language models.
In general, Debezium by itself should never miss any event. If it does, that’s considered a blocker bug which the development team will address with highest priority. After all, Debezium’s semantics are at-least-once (i.e. duplicate events may occur, specifically after an unclean connector shut-down), not at-most-once.
Technologists and quants who can code in OCaml are to be found everywhere from Citi to Morgan Stanley, JPMorgan and Goldman Sachs. Bloomberg uses OCaml for an advanced financial derivatives risk management application.
We need to look to history. Techno-feudalism, I hope, can be tamed by a digital republicanism. One that accepts the reality of power but makes it non-arbitrary and ultimately controlled by the people it effects. The first step here is the reclamation of personal data and control over its use. Just as citizens in the Renaissance republics denied the great feudal lords’ control over their persons, we must deny the great techno-feudal lords’ control over our digital persons.
Book of the Week
Do you have any more links our community should read? Feel free to post them on the comments.
Have a nice week. 😉