More from Hixie's Natural Log
Over the past few months, I've been putting together some notes for future creators of UI frameworks, based on my experiences of the past 25-30 years. The work was originally commissioned by a customer for their use, but it seems like it might be of wider interest so I've decided to publish it. The primary copy is a Google doc, available at https://software.hixie.ch/ui-frameworks, on which people can leave comments. There is also a canonical PDF copy of this first edition of the document if that is a more convenient format for anyone. As always, I welcome feedback. The most helpful way to leave comments is on the Google doc itself.
When you build something, you have to pick some design goals and priorities. Ideally you do so explicitly, but even if you don't, you're still implicitly doing so based on your design choices. These choices are trade-offs. If you want to write a quiet song, it won't be loud. If you are writing a software tool and you want to prioritize speed over simplicity, then it won't be as simple as if you'd prioritized simplicity over speed. There are two main signs that you've succeeded at your goals. The first, and more pleasant, is that you get compliments about how your thing is like you wanted it to be. "I love that song, it's so quiet!" "Your tool is so fast!" Why thank you, that's exactly what I was going for. The second sign, though, is that you will get complaints. Specifically, people will complain that your thing does not achieve the things you didn't set out to achieve. "I wish this song was louder", "this tool is so hard to use". That you are receiving complaints at all means that people are aware of your creation; that they are complaining about what you specifically set out to make a non-goal is a side-effect of the fact that you made that trade-off. The worst thing to do, when you receive such complaints, is take them to heart and try to fix them. This is because by definition you wanted these complaints. They are a sign that the thing you built is built as you wanted to build it. The people complaining want something different, they don't want your thing. It's just that your thing is so good that it's the thing they're compelled towards even though it doesn't prioritize the things they care about most. If you try to fix these complaints, you will, again by definition, be compromising on your goals. If you make the song have a loud part, then it's no longer a quiet song. You wanted a quiet song. Now it's a song that's quiet in parts and loud in parts. It probably still doesn't satisfy the needs of the people who want a loud song, and now it also doesn't satisfy the needs of the people who wanted your original quiet song. If you make your tool easier to use by compromising on the speed, then now you have a tool that's both not as fast as it could be and not as usable as it could have been if you'd started with that as a goal. It's important, therefore, to separate out complaints into those that are complaints you expect based on your design goals (which you should acknowledge but not fix), vs complaints that are either orthogonal to your goals (which you can fix without compromising your goals), or that are in line with your goals (which you should prioritize, since that's what you said to yourself was most important in the first place).
My involvement in web standards started with the CSS working group. One of the things that we struggled with as a working group was that we would specify how the technology should work, but the browser vendors' implementations weren't exactly what we intended, and web authors would then write web pages that worked with those browsers, even though that meant the web pages themselves were also not doing things like the specifications said they should. The folks I worked with at the W3C (especially the academics and people working for organizations that did not themselves implement browsers) would frequently bemoan this state of affairs, expressing surprise at how they, the people in charge of the standards, were not being respected by the people implementing the standards. One of the key insights I had very early on in my work, before working on HTML5, which really influenced the WHATWG and its work, is the realization that the power dynamics at work were not at all the power dynamics that the folks at the W3C described. The reality of the situation was that the power lay entirely in the hands of the users. The users chose browsers. A browser vendor that ignored what the users wanted would lose market share. Market share is everything in this space. Browser vendors want users because they can convert users into dollars (in various ways, but they typically boil down to someone showing them ads and paying the browser vendors for the privilege). In turn, the browser vendors had more power than the specifications. What they implement is, by definition, what the technology is. The specification can say in absolute clarity that the keyword "marigold" should look yellow, but if a browser vendor makes it look red, then no web author is going to use it to mean yellow, and many will use it to mean red. There is a feedback loop here: if one browser implements "marigold" to mean red, and some important web site (or many unimportant web sites) rely on it, and say something like "best viewed in ThisOrThat browser!" because that's the one they use and in that browser it looks red and red is what looks best, then the other browser vendors are incentivised to make sure that the web page looks good in their browser too. Regardless of what the specification says, therefore, they are going to make "marigold" look red and not yellow. When I realized this, I also realized a corollary: if you have two competing specifications that both claim to define the same technology, but one matches what the browsers already do while the other one does not, the browser vendors are going to find it more useful to follow the one that matches what they do. This is because they can trust that implementing that specification will get them more market share. It means they won't have to stop and think at every step, "will following this specification cause me to lose users?". It is easier for them to use a specification that takes into account their needs in this way. We actually tried to explain this to the W3C membership. There was a big meeting in 2004 at Adobe in San Jose, the "W3C Workshop on Web Applications and Compound Documents". We tried to convey the above (I didn't quite understand it in the stark "power dynamic" terms yet, or at least, I didn't really express it in those terms, but if you read our position paper you can see this insight starting to crystalize). At this meeting, we made a pitch for the W3C to continue to maintain HTML and to care about what the browser vendors wanted. Representatives from Microsoft and Sun (in many ways arch enemies at the time) supported us. I seem to recall Apple being more quiet about it at the meeting but also essentially supporting the principles. The W3C membership resoundly rejected this whole concept. One of the W3C staff even explicitly said something along the lines of "if you want to do this you should do it elsewhere". That's what led to the WHATWG being founded a few weeks later. The WHATWG was founded on this core principle — the specifications need to actually specify reality. When the browsers disagree with the spec, the spec is by definition incorrect and needs to change, regardless of how much technically superior the design in the spec is. Naturally, when you provide browser vendors with something that valuable, they will follow. You end up with a weird inverted power dynamic. The spec writer (when they follow this principle) has all the power, but only within the space that the browser vendors are themselves willing to play; and the browser vendors have all the power, but only within the space that the users are willing to put up with. It's very easy to appear to be in control when you tell people to do the thing they were going to do anyway (or at least, one of the things they were willing to do if they were to think about it). There is a (probably apocryphal) quote supposedly by Alexandre Auguste Ledru-Rollin that is often cited in mockery of bad leadership, but that perfectly matches the power dynamic here: "There go my people; I must find out where they are going so I can lead them". (Thanks to Leonard Damhorst for prompting me to write this post.)
I often get asked how many people contribute to Flutter. It's a hard question to answer because "contribute" is a very vague concept. There's tens of thousands of packages on pub.dev, all of which are written by contributors to the community. There's over 100,000 of issues filed in our issue database, filed by more than 35,000 people over the years (the exact number is hard to pin down because people sometimes delete their GitHub accounts; about 700 issues have been filed by people who have since deleted their account). Many more people still have used the "thumbs-up" reaction to indicate that an issue matters to them, with almost 165,000 thumbs up from about 45,000 people. All of these people are valuable contributors to Flutter. Usually, when pressed, people try to clarify by asking about "the core team". Again though it's hard to say exactly what that means, but let's assume they mean "people with commit access". That is, people we trust enough to have added to the GitHub repo as collaborators. This includes people who work on Flutter for companies like Google, Canonical, or Nevercode, and it includes people like me who are self-employed and/or contribute to Flutter on a volunteer basis. Currently that's about 280 people. So is that the answer? Well, no, not really. Some people have commit access but aren't active (maybe they got access because of their employer, but were then reassigned to work on another project, and the bureaucracy hasn't caught up with them yet — we only audit the membership occasionally because it's rather tedious to do). Some people have been very active recently but don't have commit access (e.g. because they were just laid off and a bot automatically removed their access; they might even resume working on Flutter in the future, as a volunteer or funded by another company). So what's the answer? I recently drilled down through our data to see if I could answer this. I will caveat the following numbers by saying that this changes all the time. We added a new team member just today (hi Nate!) who is not counted as a team member in the following numbers because we collected the data a few weeks ago (it takes literally days to scrape all the data from GitHub, and then hours to explore the resulting very large and very slow spreadsheet). Also, some of my definitions are a bit arbitrary, and slightly tweaking the limits would probably change the numbers noticeably. First, I collected a list of everyone who has ever created an issue, commented on an issue, put an emoji reaction on the first comment of an issue, or submitted a PR, excluding bots and people who deleted their GitHub account. (Actually Piinks did the actual data collection. Thanks!) I limited this to a subset of the GitHub repos of the flutter org that is relatively inclusive but does not count everything (we have a lot of historical repositories and so forth). This finds about 94,357 people. (So there you go. The Flutter team is about a hundred thousand people!) To avoid padding the numbers with people who left the project long ago, and to avoid counting "drive-by" contributors who came, did a bunch of work, and then left, I then limited the data set to people who contributed over a period of more than 180 days, and who last contributed sometime in 2024. Because of the definition of "contributed" described above, that means that someone who added a thumbs-up to an issue in December 2020 and then filed an issue in January 2024, and did nothing else, is included, but someone who submitted two PRs in March 2024 is not. Like I said, this is a bit arbitrary. Anyway, that leaves 3,839 people, of which 182 currently have commit access, 27 once had commit access but don't currently (these are mainly people who either got laid off recently and had their commit access revoked by an automated process, or people who were once team members, left, lost access from inactivity long ago, and then later came to comment on issues or file new issues — it's surprisingly common for people who once worked on Flutter full time to stick around even when their employment changes), and about 3,627 people who have never had commit access. Of those who have never had commit access, 2,407 have filed at least one issue or submitted at least one PR (accounting for a total of 12,383 issues and 2,613 PRs). Of those, 341 have filed 5 to 9 issues (2,242 issues total), and 296 have filed 10 or more issues in their lifetime (7,021 total issues). Similarly, of the "never had commit access" cohort, 73 people have sent 5 to 9 pull requests in their lifetime (458 total PRs) and 47 have sent 10 or more (1,321 PRs total). (For context, 4,663 people have ever submitted a pull request, and 429 have ever submitted more than 10 PRs.) Of the people who currently have commit access, 98 people have submitted more than one PR every 3 weeks on average since they first got involved (accounting for 49,173 PRs), 75 people have closed at least one issue every 3 weeks (accounting for 48,490 total issue closures), of which 10 are not in the first group (mostly that's our triage team), and 150 people have commented at least once every 3 weeks. A follow-up question a lot of people ask is "do they all work for Google?". This is a surprisingly hard question to answer. There are a lot of weird edge cases. For example, one person worked on Flutter for a company that Google hired to work on Flutter, but then quit that company, asked for their commit privileges to be removed, but continued to be active in the community. Several people who have quit Google (such as myself), or been laid off by Google, have continued to be active in one sense or another (I think I submit more code to Flutter now than I did in my last year at Google). It's also hard to answer because a lot more people at Google contribute to Flutter than just those on Google's Flutter team, and a lot of people on Google's Flutter team contribute in ways that don't show up on GitHub (e.g. product management, marketing, developer relations, internal tooling). Of the 98 people who have commit access, have been active for more than 180 days, have contributed at least once this year, and have submitted more than one PR every 3 weeks on average for the entire time they've been contributing, I estimate (based on what I know of people's employment and so forth) that about 85% are Googlers or somehow get their funding from Google, and about 15% are currently independent of Google. (This is by no means the entirety of the Google team contributing to Flutter; as I mentioned earlier, many folks at Google working on Flutter don't appear in these statistics.) I'm not sure what conclusion to draw from this; it's both more people than I expected to see funded by Google, which is great, and fewer people that aren't funded by Google, which is less great. On the other hand, it's still a significant number of non-Google-funded people. Is it enough? I think that really depends on what your goals are. I think if your goal is for Flutter to be an order of magnitude better than other UI frameworks, then frankly no, it's not enough. There is a ton of work to be done to get there. We know what it would take, but we don't have the people to do it today. On the other hand if your goal is to be a great framework, on par with others, then it's probably adequate. It would certainly be difficult to continue to be great with fewer people today. Of course, that may change as we complete big efforts, or as we take on new ones, or as the landscape changes, it's all hard to predict. That said, I would love to see more direct contributions from non-Google sources, if for no other reason but to end this silly "will Google cancel Flutter" line of questioning that has followed the project since its inception. It's a dumb question. Flutter's an open source UI framework. It will never die. It will become old and something else will shine brighter one day, just as happens with literally every other UI framework ever. That's just how our industry works. There's no reason to believe that'll happen any time soon though, and certainly no reason for it to happen earlier for Flutter than any other modern UI framework.
More in programming
Five days and 540km on a bike across Kyushu, without a laptop. Notes on the peace of riding all day, convenience store dinners, and Japan's emptying countryside.
I found a wonderful free crochet pattern for the Pokémon Ditto from Loops by Kat, but I decided I wanted to make a version with more than 3 "leg" blobs like the original pattern had. Specifically, I wanted to add two more smaller blobs in the back to help stabilize Ditto when they're sitting and also add a butt to our favorite transforming blob. Hazel Knits Cadence, a worsted weight superwash merino, in Terrywinkle that I had in my stash (I used about a third of the skein/36 grams/66 yards/60 meters), and I crocheted with a 2.5 mm hook. I make very tightly stitched amigurumi - do whatever is a good fit for you and your yarn! I used two 6 mm round safety eyes, and I embroidered the mouth with scraps of Hazelknits Cadence in Coral, Bloom, and Signature Black. The above photo is more true to color than some of the ones I took where Ditto looks more pink. I made Ditto's 2 additional booty blobs intentionally off-center by embracing the natural lean of crochet stitches - this gives a more animated or "in motion" feel to Ditto. As an added bonus, this simplifies the pattern as everyone's tension will be a little different, and thus everyone's off-centering a little delightfully different, too, but centering it instead wouldn't be the same for everyone! Here's all the information to modify Kat's crochet pattern to create a 5-blobbed Ditto. I use US crochet terminology. This pattern modification is also available as a pdf. Follow Kat's pattern through round 13. At round 14, it's time to start modifying to add the space and shaping for the additional blobs. There are also 3 more rows in this section to better fit the new 5-blob proportions. Modified increases in body, beginning at round 14 (with additional rounds, too): Round 14: 3 sc, 1 inc, 46 sc, 1 inc, 3 sc (56) Round 15: 56 sc (56) Round 16: 2 sc, 1 inc, 2 sc, 1 inc, 44 sc, 1 inc, 2 sc, 1 inc, 2 sc (60) Round 17: 60 sc (60) Round 18: inc, 4 sc, inc, 48 sc, inc, 5 sc (63) Round 19: 63 sc (63) Round 20: dec (through front loop of first stitch and both loops of second stitch), inc, 9 sc, inc, 38 sc, inc, 11 sc (65) At this point, the original pattern suggests placing the safety eyes, but I think it makes more sense to do this after you've made the "butt" and divided the stitches the 3 main "front" "leg" blobs from the original pattern. You'll still have plenty of access to place them then. Roadmap for doing 5 "leg" blobs instead of 3 To make 5 blobs, you use some of the main body's stitches and some created stitches for each. Created stitches are a combination of chains, pickups on the back of chains, pickups in the connections between the chains and the main body, and pickups in the connections between the chains. The "butt" "leg"/mounds are created before the 3 original "legs," and the left "butt" "leg"/mound is made before the right one. 1. Left "butt" "leg"/mound: This is the left "butt" cheek when viewed from the bottom. (While this is really the right butt cheek from Ditto's perspective, I'm calling it left because we're crocheting while viewing from below so it's backwards.) The start of round is the same as the start of round from making the body of the blob, so you can continue as usual. Round 1: 13 sc, chain 11 (24) Round 2: connect chain at previous start of round (skipping 52 stitches) then 24 sc (24) Round 3: (6 sc, dec) x 3 (21) Round 4: (2 sc, dec, 3 sc) x 3 (18) Round 5: (sc, dec) x 6 (12) Round 6: 6 dec (6) Finish off by looping through front loops of all 6 stitches and pull tight, etc. 2. Right "butt" "leg"/mound: This is the right "butt" cheek when viewed from the bottom. The start of round for this "butt" "leg" is 12 stitches before the body/left "butt" leg start of round. This is the second stitch in an increase from the last round of the body. Round 1: 12 sc, 1 dec in the corner of body connected to first "butt" "leg" (pictured below), 4 sc in the bottom loops in the existing chain, chain 7 (24) Round 2: connect chain at start of round for this "butt" "leg" (skipping other blob's additional 7 chains and 40 stitches in the main body) then 24 sc (24) Round 3: (6 sc, dec) x 3 (21) Round 4: (2 sc, dec, 3 sc) x 3 (18) Round 5: (sc, dec) x 6 (12) Round 6: 6 dec (6) Finish off by looping through front loops of all 6 stitches and pull tight, etc. Getting back to the original pattern for the 3 main "legs": After both "butt" "legs"/mounds are done, there are 40 stitches on the original body, 7 chain stitches that can be picked up from each "butt" leg" that haven't been picked up in a second location, which totals the original 54 stitches - this sets you up for the 3 blobs as described in the original pattern. Between each "butt" "leg" and the main body and between the two "butt" "legs" themselves are an additional 3 connections that are also picked up and decreased appropriately to keep the seams between "legs" tight. As a reminder, here's the road map, which shows how to divide up these stitches for the Ditto's main three "legs:" You might find it helpful to visualize this division by temporarily stringing up some scrap yarn to preview the divisions: If you'd like, place stitch markers like I did to delineate the division: Now that you've visualized how you're dividing the main "leg" blobs, it's a good time to place safety eyes if you're using safety eyes for eyes. I placed mine between rows 7 and 8 counted down from the small head blob and left 8 stitches between the eyes. Now you can proceed to make the main legs in the same order as the original pattern. 3. Main "leg" blob 1: This is the left "leg" blob labeled 1 in the original pattern. To start, connect your yarn into the first stitch in the main body after where the main body meets the left "butt" mound. Round 1: 18 sc, chain 9, 1 sc in the 2nd bottom loop in the chain right of the body/left "butt" mound connection, dec through the leftmost bottom loop in the chain on the left "butt" mound and the loop on top of the meeting of the main body/left "butt" mound (pictured below). (29) In the start of the original pattern "leg" 1, here are the two loops in the decrease through the left most chain loop on the left "butt" mound and the loop on top of the meeting of the main body/left "butt" mound. Follow the original pattern's Rounds 19 to 23 and finish off blob "leg" 1 as written. 4. Main "leg" blob 2: This is the right "leg" blob labeled 2 in the original pattern. To start, connect your yarn into 15th stitch from where the main body meets the right "butt" mound - there should be 7 skipped stitches along the front of the Ditto between this start of round and the left "leg" blob's last stitch on the front side of Ditto before the chaining. Round 1: 15 sc, dec through the loop on top of the meeting of the main body/right "butt" mound and the rightmost bottom loop in the chain on the right "butt" mound (pictured below), 4 sc in the bottom loops in the chain on the right "butt" mound, chain 9 (29) In the start of the original pattern "leg" 2, here are the two loops in the decrease through the loop on top of the meeting of the main body/right "butt" mound and the rightmost chain loop on the right "butt" mound. Connect chain at start of round for this blob "leg" at the beginning of round 2's first sc and continue the original pattern's Rounds 2 to 6 for blob "leg" 2 as written. As in the original pattern, now is a good time to stuff your Ditto most of the way! 5. Main "leg" blob 3: This is the center "leg" blob labeled 3 in the original pattern. Round 1: Start in the corner next to left "leg" 1 on the front side (pictured below). 37 sc into the following stitches you pick up: 7 stitches along the main body 1 stitch in the corner of the main body and right "leg" 2 9 stitches in the bottom loops of the chain along right "leg" 2 1 stitch in the corner of right "leg" 2 and the right "butt" mound 2 stitches in the bottom loops of the chain along the right "butt" mound 1 stitch in the corner between the two "butt" mounds 5 stitches in the bottom loops of the chain along the left "butt" mound 1 stitch in the corner of the left "butt" mound and left "leg" 1 9 stitches in the bottom loops of the chain along left "leg" 1 1 stitch in the corner of left "leg" 1 and the main body (37) My lime green running stitch marker indicates the start of round pickup location for making the main blob "leg" 3. To compensate for the extra 5 stitches from those corners, I decreased in the second round at the corresponding locations instead of just single crocheting the whole way around to ensure the end of my round 2 of this section had the specified 32 stitches: Round 2: 6 sc, dec, 8 sc, dec, sc, dec, 4 sc, dec, 8 sc, dec (32) After round 2, I continued with the rest of the original pattern to finish the middle blob "leg" 3. Make the two arms according to the original pattern. I attached them less symmetrically than the original suggested. Finally, embroider a mouth for Ditto. You can either do this in black to match the safety eyes or create a two-toned mouth in red and pink to make a very happy looking Ditto like I did. FYI, this is also documented as a project on Ravelry (though it's a lot harder to follow since photos couldn't be inline).
Being focused is about doing less things, and giving more space to the stuff that matters. I am spread out on too many projects and I can’t give enough attention to all of them. It’s time to clearly sort out which one I will keep maintaining and which I should drop entirely. Chuttersnap Mikochi Mikochi is a self-hostable remote file manager I’ve started in 2023. It’s my most successful open source project. I personally use it every day, and thousands of people have adopted it and started using it on their own servers. This year the project has been receiving more and more attention. Maybe it just reached a critical mass of users. In practice, that means a bit more maintenance work, since people have been pointing out bugs and suggesting new features. At the same time, working on Mikochi is fairly enjoyable. It’s a small, self-contained project made with technologies I enjoy working with. I’ve got no complaints about spending some time working on it. Verdict: Keep maintaining Mikochi and actively continue improving it. Itako Itako is an investment portfolio tracking SaaS. I built it for the most part between 2023 and 2025. In terms of scope, it’s the largest project I’ve done entirely by myself. Is it a success? Not really. There are definitely a lot of people who tried Itako, and some who’ve kept using it a bit. But it’s not engaging enough for anyone to consider getting the premium plan. And honestly I don’t find myself using it that much. My view of investing and portfolio management has significantly shifted since I started this project. While I used to think that more sophisticated analysis would lead to better returns, I now realize that the only strategies that consistently make money are honestly stupidly simple. Or maybe it’s a U curve: on one end you’ve got a monkey throwing darts, on the other end you’ve got Jane Street, and everyone in-between fails to beat the market. Anyway, the biggest issue with Itako is how dull working on it is. The interesting parts of the projects are working on statistics and data visualization. But in practice, 90% of the time is spent tediously fixing data, understanding why some German stocks have an exchange labelled XETRA instead of XETR, why some Belgian stocks still have quotes while Euronext shows them as delisted… Verdict: Shut down Itako today. Milou Milou is static site generator for press kits. It started as a fork of presskit.html which was completely abandoned at the time. It’s a very useful tool. At Tsukumogami, it’s part of our Cloudflare pipeline which builds our website in a few minutes whenever we need to change something. It seems like a few other game studios have picked it up, although I rarely get any feedback about it. I’ve not spent too much time working on it in 2026. I think it could be interesting to come back to it and add a feature or two. Better support for multiple languages could be an idea. At least, I need to make sure we don’t fall into the good old Node.js dependency rot. Verdict: Spend some time on Milou. Luluka Earlier this year, I created Luluka. It’s a CLI that allows previewing the output of Ebitengine shaders. I do still think Ebitengine is a great engine and Go is better suited for game development than most people realize. However, after quite a long period of experimentation around it, we’ve decided to remain a Godot studio. On top of having quite a large sunk cost in infrastructure surrounding Godot, the tooling and possibility of doing 3D games still makes it a more attractive engine for our use cases. And I’ve been doing a lot of 3D work lately, so better put those hours to use. Verdict: Archive the project. The rest I’ve had a lot of good ideas that stayed as almost empty repositories in my GitHub. For example, Hakumei was a terminal companion to Mikochi. I made a few commits on this project and never found the time to work on it again. I have to admit I’m probably never going to have the time to work on all those repositories. Time to archive them all!
This post was previously published through my newsletter on November 9, 2026. From time to time, I will republish newsletter issues on my blog here if I think people (and search engines) might benefit from it. If you want to read everything I’ve written once it is posted, I recommend signing up for the newsletter. If you’re doing anything to stay on top of what is happening in the world of software testing and test automation, like browsing or even posting on LinkedIn, which I stopped doing for the foreseeable future, or reading blog posts - my preferred way of keeping up these days, and I highly recommend Dawid Dylowicz’s excellent Software Testing Weekly aggregator mailing list if you prefer longer-form reading, too you’d be thinking that AI has consumed our entire field, and that everybody (or at least those who post or write in public) is an AI expert these days. The reality, though, is far from that. Yes, some people use a lot of AI, even when I would only call a small fraction of them ‘experts’. I definitely do not classify myself as an ‘AI expert’, by the way. A much larger group uses AI on a less frequent basis, and they’re getting some value out of it. Sometimes more, sometimes less. That’s the group I’m probably in. And then there’s a group of people who haven’t really been using a lot of AI at all so far, beyond maybe the odd prompt to ChatGPT or Claude. I don’t have the actual numbers, but I suspect that this group is a lot larger than you would think from browsing your feed or surfing the blogosphere. You just don’t see them, because they’re typically not the people who do a lot of work in public. A few weeks ago, I started working with a small group of testers to teach them how to use AI in their daily work. This work covers both test automation as well as broader software testing tasks, as some of them contribute to automation on a daily basis, but not all of them do. I’m working with them one day per week, on average, for three months, and we’re making small steps towards making AI a more prominent tool in their tool belt. So, what are we working on right now? I’m pulling ideas from my ‘Valuable Feedback, Fast with AI’ course, and so far, we’ve spent some time setting up Claude, going over some important concepts such as context windows, models, tokens, and (of course!) the probabilistic nature of LLMs, and what that means for the results they provide. We have gone through several initial conversations, using Claude to learn more about and review specific user stories, which was an excellent segue into using talking about MCPs and setting up a connection to both Jira and Confluence using the Atlassian MCP. Last week, we also started experimenting with writing straightforward skills to make it easier for Claude to repeat certain actions in a structured way. These actions ranged from reviewing stories and acceptance criteria to refactoring small chunks of code in their (Robot Framework-based) test automation suite. What I loved about this is the avalanche of ideas that popped up in their heads when they saw how easy it was to use Claude. We made a short to-do list of ideas to try out for the next couple of weeks, and I’m looking forward to seeing them getting more confident around using Claude. One thing I have noticed is that their level of comfort around using AI varies quite a bit. Some of them are naturally quite cautious, which means they’ll ask ‘can we really do this?’ and ‘should we really do this?’ frequently before even submitting a single prompt. This is a good habit, I think, but we’ll have to do some work to get them more comfortable with using LLMs, without losing too much of that caution. Others in the group are the opposite: I need to hold them back a little, because otherwise they’ll create a beautiful mess within days, simply because they see so many opportunities to use Claude in their work. That, too, isn’t inherently bad, but I do want to teach them not just how to get the most out of an LLM, but also (more so) how to use AI prudently. Over the next couple of months, we’ll dive deeper into writing better prompts, creating and structuring skills, better context management, and creating and using agents, and probably a few more things, too. I’m looking forward to seeing these people grow more comfortable around using AI in their daily work, and figuring out where AI genuinely adds value and where it doesn’t in their own context. Is there a deeper message in this newsletter? Maybe not, and I’m just happy to tell you about this project, and how it is a great example of a what I call ‘corporate mentoring’, a very efficient way of teaching a small group of engineers new skills, within their own context. But maybe it’s also to show you that, despite what you might see on the Internet, especially those places featuring a lot of self-congratulation (I’m looking at you, LinkedIn), not everybody out there is an AI expert. At least not yet. So, if you feel like you’re ‘behind’ simply because you’re not running at least 17 agents continuously, creating 5400 merge requests per day, know that you’re not alone. Quite far from it, I would say.