Profile avatar
kitmacleod.bsky.social
Environmental researcher 🌍 | data engineer | software engineer Digital Tool Specialist @ UK Centre for Ecology & Hydrology https://www.ceh.ac.uk/ Own views https://bit.ly/KitMacleod - Google Scholar https://bit.ly/MonitoringEarth - Bsky feed
637 posts 1,019 followers 344 following
Prolific Poster

Interesting arguments why DuckDB should be considered a protocol. I am just sorry that some poor duck had its feet roasted in the process.

Finally, we've created deep dives for every single episode! Read about the process and how they fit with the podcast. cc @mkennedy.codes #python talkpython.fm/blog/posts/d...

My family has several generations of talented artists (I am not, yet, one of them). I like to practice when I am in beautiful settings like these.

🆕 #CNG2025 talk now available In this presentation, Matthias talks about the efforts to standardize satellite imagery, making it more readily available and usable for immediate analysis. youtu.be/vj4j4lJNyk0

And a fun game...

Including a picnic by the Thames.

Fun team development day at @ukceh.bsky.social with @mattfry-ukceh.bsky.social

Good to see investment in UK digital skills www.gov.uk/government/n...

“C3S is not just about access to free and open climate data. It is also about the story we tell about these data. The visual narratives are a key part of this storytelling.” @carlobuontempo.bsky.social Yes,access and storytelling inform, inspire and enable all. climate.copernicus.eu/role-data-vi...

"Big changes are coming to the ways you can access #USGS water data in 2025. A number of our legacy products and services will be turned off and we’re excited to unveil the next generation APIs and user interface of the USGS Water Data for the Nation ecosystem." waterdata.usgs.gov/blog/wdfn-up...

"Our perspective on how remote sensing and citizen science can be combined through machine learning to provide high quality, near real-time reporting on habitat condition. This could be transformative for environmental reporting." www.ceh.ac.uk/news-and-med... @ukceh.bsky.social

uv is one year old today. Hard to believe its only been a year. The growth, adoption, and impact surpassed my wildest expectations. Happy birthday, uv! 🥳

Want to easily use LLMs from R? ellmer 0.2.0 provides: • Batch & parallel processing • A universal parameter interface • Cost tracking • Support for Mistral, Hugging Face, & Portkey And much more! Read all about it at: www.tidyverse.org/blog/2025/05... #rstats

Great Tables is the #Python library for creating display-ready tables from your #Polars DataFrames and more. In their 3-part video series, @mchow.com & @richmeister.bsky.social show how to add structure, formatting, & styling to go beyond the default. Explore the details: posit.co/blog/level-u...

May 2025 experienced the first 430 ppm+ monthly #co2 average in human history, per Keeling Curve: scripps.ucsd.edu/news/annual-...

To #SaveOurOcean we must conserve, sustainably manage and restore marine and coastal ecosystems, including deep-sea ecosystems. At #UNOC Ocean Action Panel, underscored the importance of integrated policy and management solutions, based on strong science, to protect marine and coastal environments.

"I’m discussing only the implications of LLMs for software development. For art, music, and writing? I got nothing. Im inclined to believe the skeptics in those fields. I just don’t believe them about mine..It’s getting the same kind of attention that smart phones got in 2008" fly.io/blog/youre-a...

Joe talking about the need to be aware of what LLMs are good and bad at, as there is a "tidal wave of change that is ahead of us" Harnessing LLMs for Data Analysis | Led by Joe Cheng, CTO at Posit - YouTube www.youtube.com/watch?v=owDd...

Handy to know relative strengths of progressive web apps and native apps thenewstack.io/when-to-use-...

How Java has evolved and future plans thenewstack.io/java-at-30-j...

If you are interested in what has happened in LLM space over the past 6 months - listen to @simonwillison.net www.youtube.com/live/z4zXicO...

"Spaces let you ground Copilot’s knowledge in a curated set of specific code, documents, notes, and more. With this extra context, Copilot becomes an expert in the task at hand—from understanding how a system works, to why it was built in a particular way.." github.blog/changelog/20...

@swyx.io useful steps for producing AI news and more generally for AI intensive applications. AI Engineer World's Fair 2025 - Day 1 Keynotes & MCP track ft. Anthropic MCP team - YouTube www.youtube.com/watch?v=z4zX...

Huge research effort, that is well documented. Not 100% sure of added value compared to GBIF - why multiple species occurence databases (I may have missed some key reason). biotime.st-andrews.ac.uk

Thanks, for sharing this useful summary.

New @crew-waters.bsky.social call for proposals to apply behavioural science evidence to produce a household flood plan template and consider future actions to support uptake. Deadline to apply: 9th July, midday. Full details here: tinyurl.com/mrxvphe2 FAQs: tinyurl.com/2j2yappm

@element84.com geospatial tech radar is a great community resource- made even better by your input.

Amulya is a star (a great colleague)- well done @amulyachevuturi.bsky.social

🌍 A new @nature.com study co-authored by UKCEH’s Prof Chris Evans reveals ancient, buried carbon—locked in soils and rocks for millennia—is finding its way back to the atmosphere via global river systems. 🔗 www.nature.com/articles/s41... #ClimateScience #CarbonCycle 🧪 1/

An exciting job opportunity to work with my amazing ecological colleagues- they have a huge amount of ecological monitoring and data science experience

"When you ask an AI model to help you analyze some data, it can suggest approaches+even write code, but it can't actually execute that code or interact with your data directly. An AI agent,on the other hand,can actively work with your data: loading files,running analyses, generating visualizations."

Excellent advice on prompting including providing rich context and being specific.

Catch up with Simon's and many more key talks on AI engineering.

Lucy Barker from @ukceh.bsky.social presenting the current status of the Water Service at the #CopernicusClimate General Assembly — great work, Lucy!

Key route map, highlights need for better data and evidence: "challenges of managing soil systems across multiple land uses and policy themes and to develop an overarching strategy for delivering improved soil security across Scottish landscape"

🌍 From in-depth climate analysis to monitoring the air we breathe, #CopernicusClimate and #CopernicusAtmosphere help safeguard our environment. The services’ high-quality, open and free data inform decisions for a more resilient, healthier world. #WorldEnvironmentDay

Three charts based on #C3S data you shouldn’t miss this month: 📈 April 2025 was the second warmest April globally on record – #C3S 1️⃣ bsky.app/profile/cope... Dive into the bulletin: climate.copernicus.eu/surface-air-...

2025 Water Resilience Strategy focuses on three objectives: Restore and protect the water cycle as the basis for water supply; Build a water-smart economy to boost competitiveness, attract investment; Secure clean and affordable water and sanitation for all. environment.ec.europa.eu/topics/water...

Fascinating age group shift from 50-54 to 18-24 in the space of 6 years (and COVID). Putting a huge (and unnecessary) strain on the many volunteers who put themselves in danger to rescue/help others. Be prepared. ourhttps://www.ordnancesurvey.co.uk/news/18-to-24-biggest-age-group-in-need-of-rescuing

Thank you all (1k) for following me - appreciated 😀. Please feel free to say what types of content/tools/techology you are most interested in.

Thx @opencholmes.bsky.social: Shared geospatial data is key to address challenges, we need a much more complete understanding of what is happening in the world than exists today. And to get that understanding we need the right ecosystem for the creation of that data+tools to make the data accessible

Adoption of common schemas across orgs can accelerate the people, data, models feedback loop

Components of an GSDE AI ecosystem: benchmark labeled data; metrics; models; tools to generate model outputs; tools to expand/improve the benchmark data.

Need robust metrics about things users care about

A classic keynote, a few key points: if key geospatial datasets were in Iceberg format then the friction to bring data inside a company goes to zero...and they can access data about the state of our planet

Great news...

V. Useful: "Just to be upfront and get it out there: no. Despite what you may have heard, Zarr is not (yet) a replacement for the Cloud-Optimized GeoTIFF (COG) format. Zarr is great for Level3 and Level4 n-dimensional data cubes. COG is great for Level1, Level2, and other non-data-cube rasters"

"For R coding tasks, we recommend using OpenAI’s o4-mini or Anthropic’s Claude Sonnet 4. OpenAI’s o3 performed the best on this evaluation but is also ten times more expensive than o4-mini and around three times more expensive than Sonnet 4."

Useful advice, from Simon, on how to improve your prompting, based on a UK government ministers use of ChatGPT.