I’m a systems thinker. A patterns thinker. A process person. A metaphor lover. A BioOptimist.
Feel free to pick just one of the above, whichever is most relatable to you. They are all ways to say the same thing: patterns appear in systems of processes, metaphor is a useful tool to quickly understand new systems, and biology is fertile ground for metaphor as it’s full of complex, interrelated systems.
Most recently, I’ve been applying this to training systems. Specifically, training systems for laboratory automation. At the simplest level, you have inputs (untrained students), a series of steps that alter the input (education and skill development), and a new output (trained students with skills).
There are endless examples of this in biology across different scales. From energy production in cells to growing corn with fertilizers, there are a lot of lessons we can learn about effective, efficient, and enduring processes. We’ll start there, and then see how this all applies to training systems.
Context is the Difference between Inefficient and Efficient
Biological processes tend to be discussed in isolation, and it’s a real shame. You generally see diagrams that show combining inputs at various steps, resulting in an output. It’s like driving down an unfamiliar highway; sometimes you’ll see on-ramps or off-ramps, but there’s no sense of where those other roads came from or go.
But just like taking someone’s words out of context changes the implied meaning, taking a biological process out of its context changes the implied efficacy and efficiency.
Let’s look at oxidative metabolism as an example. Big words aside, this process is essential to how our cells convert the food we eat into usable energy and is the reason we breathe air. No big deal? It’s a big deal.
If you look at a nutrition label, you see calories as the amount of energy contained. But at the cellular level, the name of the game is a little molecule called ATP. You were first introduced to this topic in high school biology, and if you’ve forgotten it completely, I don’t blame you. Don’t worry, here’s all you need to know about it to understand my point.
Oxidative metabolism is the process by which our bodies use oxygen to make energy. This is how you keep your muscles moving during aerobic exercise, like running a few miles. Your cells use the glucose in your body to produce ATP, the energy currency of your cells. In theory, you can get up to 38 ATP molecules from 1 glucose molecule. In reality, you get somewhere between 30-34 ATP per glucose molecule.
Why is the actual yield less, and why does it vary? Is it because biology is inherently inefficient? That’s what I thought when I was first taught this in high school, but it turns out that the loss in yield is actually because biology is spectacularly efficient.
Some of the molecules involved in oxidative metabolism are also essential in other pathways. These other pathways include fatty acid synthesis (essential for the structure of your cells), amino acid synthesis (essential building blocks of proteins), producing heme for hemoglobin (essential to move oxygen through your body), and gluconeogenesis (essential for blood sugar regulation).
At a glance, what appeared to be an inefficient process was actually a process taken out of context. Once placed in the proper context, it’s clear that the price of a few ATP from your yield is a bargain to keep your entire cell running smoothly.
Understanding the efficiency and efficacy of biological pathways is always context-dependent.
Balance is the Glue that Keeps Systems Working
Sometimes you have a process that seems to work well, but you want more of your output. To go back to the highway analogy, we’re trying to get as many cars as possible through this highway, so instead of driving them individually, we’re putting them on car carriers. This way, for every one vehicle we are getting 6 cars through.
Biologically, we’ve done something very similar when we use fertilizer to increase crop yields. Same number of plants, but more ears of corn. It sounds pretty ideal, but what actually happens?
Let’s start with a very specific case where you add only nitrogen to the soil. At first, you get exactly what you hope for, more corn. But the corn plants aren’t the only thing living in the field and impacted by that fertilizer. The microbes living in the soil also noticed the extra nitrogen and took it for themselves. This makes the nitrogen unavailable to the plants, while also fueling the microbes’ ability to break down other components of the soil.
Soon, our soil is more acidic, the organic matter is broken down, beneficial microbes are struggling to compete, and the soil loses its structure and water-holding capacity. Now you have to add more nitrogen just to maintain your crop yields, but this only further exacerbates the problem. Eventually, no amount of extra nitrogen can overcome the fact that the soil is compacted and can no longer host the microbes needed for healthy plants.
What happened? Fundamentally, you added too much of one thing, and it resulted in a degradation of your system. In our highway analogy, you forgot that car carriers weigh more than individual cars and deteriorate the road at a faster rate.
Your system is out of balance. You can indeed safely add fertilizers to a field to increase yields, but you must make other changes as well to keep the system capable of sustaining the increased production over time. Your system infrastructure must be designed and reinforced to handle the added capacity, just like the highway needs to be built and maintained differently to support a steady stream of car carriers.
There’s a second lesson to be learned from our corn field: you have to treat the system, not the symptom. Removing nitrogen does not restore the soil. Taking the car carriers off the highway does not fix the damage that’s been done to the highway.
Biological Principles to Build Systems that Last
To recap, looking at biological pathways has taught us that:
To understand a system, you must understand its context
Systems require balance between the components
Fix systems, not symptoms
These align remarkably well with some of my most salient learnings in implementing automation training. Here’s how it’s played out in practice.
Using Context to Build a Pathway to Success
I first entered the intersection of laboratory automation and education when I was working at Opentrons Labworks. I was leading the development of Opentrons for Education, an initiative developed to help bring automation into the classroom.
The common assumption is that cost is the primary barrier to automation education. But very quickly it was obvious that this was just one of many barriers, and maybe not even the primary barrier.
So I mapped out the process and barriers of implementing laboratory automation from the perspective of an educator. Here’s a simplified version of what I came up with:

When I looked at it this way, it was clear that educator know-how was the primary barrier. And it makes complete sense - how can you expect someone to teach a skill they don’t have? If we could address that barrier, we could simultaneously lower the barriers of implementation strategy and updating curricula. Paired with a smart discounting structure, we could have a real solution.
With that in mind, I developed educator training workshops. This was not a sales workshop; it was professional development for educators, teaching them how to use liquid handling robots and how to incorporate them into their curricula.
It was a huge success. Of the educators who were in attendance, over half of them were teaching with automation less than a year later.
Driving Cohesion and Collaboration with Balance
Earlier this year I had the good fortune of contributing to an NSF Engine grant1. Engine grants are unique in that they are seeking to build an economic flywheel by linking together currently disparate systems.
While working on the workforce development strategy for this grant, we faced an interesting challenge: there were strong institutions and programs to partner with, but at a glance, they all needed something different. That is, until we understood the community-to-employment pipeline.
The community-to-employment pipeline is the complete process someone goes through in their training experience, from first learning about the career path within their community, through their training experience, to becoming gainfully employed with their new skills. When we applied this framework, we suddenly had a very clear picture of what was working for each program, and the best opportunities to strengthen them.
In one case, community engagement was strong, and the training program was well-respected. But it was difficult for students to afford the mid-training internships critical for job placement. Balancing the system here came in the form of financial support for internships, effectively completing the students’ hands-on skill development and de-risking proof-of-skill for employers.
In another case, the program had a job placement rate of 100%. Wow. But they didn’t have enough students entering the program to meet the full job demand. Here, balancing the system came in the form of increasing community engagement to increase enrollment.
The System Solution that was Outside the Classroom
One of the most strikingly effective interventions I’ve seen has almost nothing to do with the training itself. A college in Los Angeles noticed that a lot of students were dropping out of several programs. When they looked into it, there was no correlation with student performance or program of study. What was going on?
Tackling this problem by the symptom would have resulted in trying to incentivize or convince students to stay in their programs. But by talking to students in more detail, they learned that the problem in fact had nothing to do with their programs. The students were struggling to make ends meet, and it’s very hard to get scheduled to work enough hours when you are also enrolled in courses. Students were leaving because they needed to work more in order to feed themselves and their families.
The school responded by opening an on-campus food bank. It was a spectacularly effective response that allowed many students to complete their courses and ultimately get jobs that paid better than those they were leaving school for.
Build Better, Build like Biology
Biology has spent about 3.7 billion years perfecting processes that thrive in reality; understanding biological principles is key for building resilient and enduring strategies. As we increasingly feel the pain points of our current approaches failing, we have two choices: treat the symptom or treat the system.
Always treat the system.
You may be thinking, “Wow, what a harrowing time to submit a grant to NSF!” And indeed it was a roller coaster. Kudos to Rein Ulijn and Alma Eva Perez Perrino for their herculean effort; it was a pleasure to witness their professionalism.