Hands-on notes from building with AI since before GPT-3: machine learning and data, then LLMs, now agents.
A few years ago at Starship, I contributed to the Data is the Specification [http://dataisspec.github.io/] manifesto. The core idea is that it's better to solve problems directly against a collection of examples, as opposed to trying to generalise the problem first and then solving the general problem. Typically you would run an employee engagement survey,
What do people mean by “artificial intelligence”? When I say “AI” I mean one of two things. First it could be the experience of a product feeling intelligent. Or I could refer to what it looks like technically: that the system uses machine learning, deep neural networks, code where you optimise parameters against a dataset. The not explicitly agreed but
A startup’s AI work often starts from a developer hacking algorithms on the side. When a generalist engineer with no data science background works on prediction problems, they often don’t use a dataset at all, or at best a small one. For example, in the early days of Veriff, a front-end developer tried to build a feature
devclub.lv meetup 07.11.2019ML at Veriff: What is possible in a year?Google Docs
You can quickly iterate code, but not data. This is one of the major bottlenecks in companies trying to automate human activities, and solving it generally would change the way every machine learning team works. To understand how easy things could be, consider a simple web application. To make a small change to the site – e.g. changing the way
This post is based on my talk at North Star AI 2019 [https://aiconf.tech/]. When I speak about automation in general, and the Automation team at Veriff, I mean it in a specific sense of the word. Building software to perform typically-human tasks… whose solutions usually cannot be perfectly described in code. An example of an automation task
This is a talk I gave at a Zen of Data Teams [https://www.meetup.com/Machine-Learning-Estonia/events/260081030/] event in Tallinn. The full slides are here [https://github.com/taivop/talks/blob/master/files/loops_of_building_algorithmic_products.pdf] . My talk starts at 50:53. The two loops of building algorithmic productsThe two loops of building
Pretty much every presentation I give about AI includes the following slide, in the context of automatically authenticating users using face recognition. The idea is that even automation-reliant businesses don’t need to go 100% automatic to start serving customers: you can start with a basic model that outputs a confident “yes” (green region), a confident “no” (red region)
As the popularity of machine learning grows, so does the amount of academic papers, blog posts and software released. As an industry practicioner I need to be up-to-date with useful ones without spending hours each week combing through new developments to find the few important updates. The obvious solution to this is curation: having people or algorithms digest
Ask them if AI will kill all humans. Actually, anything along these lines will work. When will we achieve 100% automation? When are we going to build a machine that is better than humans at every kind of work? These are questions I often get at talks, interviews, or one-on-one conversations. Others associating with AI in their work
