You might have wondered how many people are behind the scenes at DOAJ and what they do. This blog post series will offer our community an opportunity to meet several of our team members and learn more about their roles and responsibilities. In this blog post, we will meet Brendan O’Connell, Platform Manager.
Hello Brendan! Tell us a little about your background before joining DOAJ!
I’m the Platform Manager for DOAJ, which means I make decisions at a high level about development priorities for the DOAJ software platform, which includes our public website and search interfaces, our APIs, and the internal journal evaluation system that powers the work of our Editors. In doing this work, I need to balance the needs of many different kinds of users: researchers looking for a high-quality open access journal to publish in, aggregator platforms and library catalogs that consume our API data, and our Editorial team. Since there’s always more that we want to do to improve our platform than we have the time or resources for, I spend a lot of time balancing competing demands and making decisions about development priorities.
I am the main point of contact with Cottage Labs, a development firm DOAJ has contracted with for over a decade. We don’t do any development in-house, so I am the primary link between the developers and the users – translating user needs into features we ask the developers to build, and translating back to DOAJ stakeholders like our leadership and Editorial team what is realistic for the developers to accomplish.
In terms of my background, I have worked in a lot of jobs that are in this vague space between people and technology. I was a librarian for a decade, and a lot of my work in libraries was around instructional technology, trying to help faculty members at my institution incorporate the library and library technology into their courses. A few years ago, I made a transition into software and was a software engineer for a few years at two other non-profit open infrastructures, Zotero y Thoth.
Your background has a lot of traces of community-led work and non-profit organisations. Has this been a conscious choice, and is that also what made you apply to this role at DOAJ?
When I decided to become a librarian, I wanted to be involved in things that had social impact. That’s very much aligned with my personal values – making the world a better place.
My dual backgrounds as a librarian and a software engineer, and my personal belief in social justice and the open access movement led me to my current role at DOAJ. I was really excited to find a job that aligned with my values and combined all the kinds of work I love to do.
You mentioned earlier that your role is like being a translator between people and technology. What do you find most challenging about being a translator between the two?
That isn’t the part of my job I find challenging, that is actually my favourite part. I’ve always enjoyed being a populariser of technology, helping people to understand technology and how they can use it to enhance their work!
What do you find the most challenging about your role?
It’s very different from my previous role as a software engineer, which had a lot of monotasking – really focusing on one development task at a time and working on it until it’s done. This role is completely the opposite, because it is all about prioritisation, both for the platform and for myself. I am constantly managing my own time while also effectively managing the time of the developers as well, because we often have ten things that we as DOAJ think are really important. However, the developers can only effectively work on one thing at a time. The challenging part for me is to figure out what is the one most important thing for the developers to work on at a given moment.
Do the priorities change or do they stay the same?
We have a pretty strong user focus, so things like bug fixes that affect a lot of users will always be bumped up on our priority list. At the moment one of the developers at Cottage Labs is pretty much allocating 100% of his time to DOAJ to work on a complete redesign of our internal Editorial system, but even so, the maximum we can really allocate for pure feature development for him is around 10-13 days a month. Because unexpected stuff comes up all the time that needs to be dealt with quickly.
A good example of this is the site stability issues we were experiencing in late 2025 as a result of massive traffic increases driven by AI scraper bots, which I wrote about in a blog post. This didn’t come out of the blue, every open access repository and infrastructure has been experiencing it for at least a year now. We had to urgently divert the developers’ time into a pretty major infrastructure upgrade so that the editorial team could do their jobs when the public site was getting slowed down. So, we cannot plan completely, because things will always come up. That’s built into my estimates as well when I prioritise work – I’m always planning for unexpected things to come up.
Speaking of scraper bots, there’s been a lot of talk around AI recently. How does AI affect your work at the moment?
I am both using AI for my work and fighting off the negative effects on our platform at the same time. I recently built a visual dashboard of DOAJ indexed journals from Latin America and Africa, using Github Copilot (an AI coding assistant) and our open journal metadata. This is called vibe coding, where you just prompt the AI to build things without doing much (or any) of the actual coding yourself. Jason Priem from OpenAlex recently did a webinar on vibe coding dashboards using OpenAlex data, which is what inspired me to do the same thing using DOAJ data.
To be honest, I try to use AI for other aspects of my work, like project management, and haven’t found it to be nearly as useful as it is for coding. There are also real limits to what you can do with vibe coding: you often end up with what’s now called “spaghetti code.” The code works, but you have no idea how it works when the AI has designed all of the code architecture, and therefore you can’t really untangle it when something inevitably breaks.
The capabilities of AI tools are changing constantly, so I also try to return to the various tools every few months to see if something that didn’t work the last time has gotten better.
What do you like most about your role?
I really like the variety. I’m not that detail-oriented, so being a software engineer did not match my best style of working. I like being at the higher level, thinking about where the platform is going and opportunities on the horizon for DOAJ. I like that I get to talk to people to work through ideas, as I am extroverted, and coding is fairly solitary. DOAJ also has a very international team, with different perspectives and people from different nationalities and cultures. That mix is really fascinating.
What are your personal views on open access?
I think the amazing thing about open access is that you have no idea who’s going to read and benefit from the research you’ve done, and that’s really incredible. You can’t imagine the reach your work is going to have. Open access is an amazing global movement, bringing research to people who otherwise would not be able to read it. I like thinking about the impact and the social justice aspect of getting research into the hands of people who need it.
I remember that when I was a librarian in early 2020, so many journals were suddenly releasing all of the critical research on COVID-19 open access to speed up the pace of scientific research and share breakthroughs as widely as possible. Like a lot of the amazing, positive social change that happened in the midst of so much suffering during the pandemic, there’s now been some counter-reaction and retrenchment. Things have gone back a little bit to the status quo – but not all the way, and more things are open access than ever before. It obviously begs the question that if the COVID crisis was important enough to give everyone access to knowledge and research that could save lives, why aren’t other global issues and crises equally important?
Even as more research is published open access than ever before, so much of the business model still relies on Article Processing Charges (APCs) that are paid by the researcher or their institution. The big commercial publishers have all concluded by now that they can’t fight the open access movement, so they’ve instead become a highly profitable part of it, while libraries, authors, and government funding agencies continue to pay. Increasing the proportion of Diamond open access publishing – research that is free to read y free to publish – is the future we’re trying to build at DOAJ.
AI is now arguably a bigger challenge to open access publishing than the large commercial publishers. Open infrastructures like DOAJ weren’t built to handle the massive volume of bot traffic to train AI models, and ironically, are most vulnerable because everything is free! We need to figure out ways to work more sustainably with AI companies so that their relationship with us is mutually beneficial, instead of primarily extractive, as it is now. I remain hopeful that there are opportunities for us in offering Premium Metadata Services, for example, so that DOAJ can be fairly compensated by the most intensive users of our metadata.
