Something I’ve learnt from 39.5 years of using email every day for work:

  1. If you are the kind of person who communicates with another person to tell them that you need to communicate with them…
  2. Then you are a dreadfully stupid person.

There, I’ve said it. Time to exhale.

Let’s put this into business terms: extremely poor ROI and low LTV.

Let’s put this into scientific terms: consider the existence of a Bayesian prior indicating how another individual is likely to be a complete flake.

You’re welcome.

Not clear on this yet? We’ll illustrate with an example.

Alice is one…


For a full day on December 14, 2020, LinkedIn sponsored a virtual workshop called Metadata Day, followed by a public online meetup called Metaspeak. View videos of the event on their website including a set of lightning talks provided by several of the speakers.

The gist is that circa 2018, the pending regulatory compliance for GDPR, CCPA, etc., drove urgency in enterprise organizations to have a better understanding and accountability about datasets and their use cases. Several firms, about a dozen, responded in strikingly similar ways — which signals that a kind of “design pattern” is emerging for how to…


Registration is open for the Metaspeak Meetup (a free online event) on December 14, 2020, hosted by LinkedIn https://metadataday2020.splashthat.com/

This is the public portion of the Metadata Day 2020 workshop, which convenes leaders from several open source projects for graph-based dataset metadata management, along with thought leaders in metadata management and dataset governance from Google, IBM, UC Berkeley, etc.

Igor Perisic, CDO @ LinkedIn, will present our welcome address, followed by two panels of expert practitioners who will summarize outcomes from our workshop earlier in the day — (1) focus on use cases, and (2) focus on production at scale…


RLlib is an open-source library in Python, based on Ray, which is used for reinforcement learning (RL). This article presents a brief tutorial about how to build custom Gym environments to use with RLlib. You can use this as a starting point for representing your own use cases to solve with reinforcement learning (RL).

Note that this article is a follow-up to:

If you haven’t read that previous article already, check it out.

Source code for this article is available at https://github.com/DerwenAI/gym_example

The material here comes from Anyscale Academy and complements the RLlib documentation. …


RLlib is an open-source library in Python, based on Ray, which is used for reinforcement learning (RL). This article provides a hands-on introduction to RLlib and reinforcement learning by working step-by-step through sample code. The material in this article, which comes from Anyscale Academy, provides a complement to the RLlib documentation and is especially intended for those who have:

  • some Python programming experience
  • some familiarity with machine learning
  • no previous work in reinforcement learning
  • no previous hands-on experience with RLlib

Key takeaways: we will compare and contrast well-known RL examples running in RLlib to explore the essential concepts and terminology…


Their community was still reeling from an incident where local police had murdered an unarmed black man. Protests grew as the public was outraged in the wake of a court decision to acquit the murderers. Then a crowd gathered in a major city as a local protest leader made a public reading of a controversial document. Afterward, protesters marched to a nearby park and pulled down the statue of a corrupt government leader portrayed as a Roman emperor.

The date was July 9, 1776. The city was NYC. The man whose death had catalyzed protests was Crispus Attucks. He’d been…


Sebastopol, California 2020-06-03

Derwen, Inc., announces a new AI service for a novel form of risk analysis, addressing business needs during a time of heightened uncertainty to augment due diligence practices, while emphasizing resilience and ethical practices.

First and foremost, our hearts go out to those who have been victims of illegal and unjust police aggressions, to their family, friends, and neighbors. And also to those who have lost loved ones during the pandemic. We support those who are actively protesting, calling out the systematic racial violence which is so deeply ingrained in the US.

Response to Urgent Business Needs

In response to urgent business needs brought on…


Open source tools and techniques for visualizing data on custom maps

Throughout the global pandemic, many people have spent lots of time viewing maps that visualize data. Important data. People who work in data science are probably seeing increased needs to work with geospatial data, especially for visualizations. There are increased needs to understand metrics about geographic regions, to analyze supply chain, make plans that take into account local conditions and rules, etc.

This article shows how to use two popular geospatial libraries in Python:

  • geopandas: extends Pandas to allow spatial operations on geometric types
  • geoplot: a high-level geospatial plotting library

The second library is especially helpful since it builds on…


Strategy notes for technology ventures during a “pseudo-jackpot” period subsequent to the COVID-19 pandemic and its economic fallout.

The following is a collection of personal notes. Most of these points compare personal reflections from previous economic crises w.r.t. the current pandemic. This is especially targeted at questions about leading an early stage technology venture.

Our world has changed. On the other side of a global crisis, where do we land? As an enormous range of possible risks and opportunities open up, which strategies are indicated? How do we prepare for a so-called “new normal” and its emerging business environments?

The…


A curated list of dashboards for tracking the COVID-19 pandemic and related factors.

Preface

The following links provide dashboards and related resources for a number of different perspectives analyzing the global COVID-19 pandemic.

Note that the more reliable sources tend to:

  • cite data sources
  • focus on per-capita estimates instead of absolute values per region
  • show confidence intervals and predictive bounds
  • condition any reported ratios of relatively sparse data by using point estimates
  • discuss external factors which aren’t represented by features in their models

OTOH, among the less reliable sources … not so much. YMMV.

Image for post
Image for post
COVID-19 deaths/mil vs. US metro population, Earth Day 2020

Analysis and Briefings

AI-based summarization and briefings are vital, given…

Paco Nathan

evil mad scientist derwen.ai/paco ; lives on an apple orchard in Ecotopia

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store