Our Code Compliance series features perspectives from industry leaders.
ABOUT
Berkley, Santa Clarita
INDUSTRY
Municipal Building
Code Enforcement
RESULTS
My name is Eric Schneiderjohn. I'm a Plan Check Engineer at WC3. I'm also Vice President for WC3. I've been working with them for about 12 years. I work internally, help train some of our teams – mostly in charge of our online academy – and a lot of it is just plan checking, working with staff, growing the next generation, and then also working with clients.
I work in-house in a lot of cities – I've worked in-house in Berkeley, San Jose, Emeryville, Santa Clara, things like that. So, staff augmentation is part of what we do, so I get to do that a lot. And The work I get to do less than I used to – that I enjoy the most – is the plan checking part of it.
Yeah. What do you love about plan check?
One of the best parts is that you are constantly getting new things. You know, you can be looking at a restaurant, and then right after you're looking at a high-rise, and then right after you're looking at something small like a bathroom remodel. So the variety – I really enjoy.
But also, when you get a big project, you kind of just get lost in it, and you're going through either lists in your mind or you're chasing things down, and it’s just the kind of getting to do the production part and then the things I enjoy most are going through the plans, finding where things are wrong, trying to add some value to it – and by value I mean either in terms of life safety or having an effect on how the design ends up coming out.
You kind of feel like your fingerprints are on it a little bit. We're not designers – we're not making the things that are out there. What we make is our comment letter – which is the tedious part – but the going through the plans is the fun, entertaining part.
Curious – in this industry, there's a lot of rigor and complexity. I think just like a lot of problems or challenges with that. What are some of the biggest challenges you see?
Good question. There are a lot of challenges. One of the biggest problems I see is that the process is somewhat backwards.
If you're a developer or owner and you want to develop a property, you purchase the property, you have carrying costs, you put up all this upfront money, you hire designers and architects, engineers, you go through the process of design this – and by the time someone like myself gets involved – there's already a lot of money in design that has gone into it. They’re down the road.
So one of the main challenges and pushbacks we see is that anything we're calling out, asking them to do, or changing often has cost implications. I think if we got involved earlier, or if there were better knowledge on the design or developer side, many of those changes could already be implemented in the design.
Some other challenges: lack of staff, lack of resources, unfunded mandates – and just growing complexity. In my old office, we used to have the 1998 UBC and it was this big. Now we have California Building Code Volume 1, Volume 2, mechanical code – it takes up this much space on a bookshelf.
Obviously, there were other codes back then too, but there's definitely been growing complexity in what's been added over the years. EIf you even go just two decades ago, there were no energy codes; a decade ago, there wasn’t CalGreen, which is essentially a sustainability code. The growing complexity means there’s more to know and more to look for, but not necessarily more time to look at it.
Why are codes necessary?
From my perspective, codes are necessary for safety, they're necessary for public health, and they're important and necessary for things like access to everything for persons of varying abilities.
In addition to that, they help with energy and sustainability, when codes are implemented properly, they’re helping ensure life safety. One thing we say in training: when we’re doing our job right, nothing happens. You know, buildings don’t collapse, you don't get children falling or places where there aren't guards, you don't have nuisance tripping on electrical, things like that. So, if you're doing it properly, it's not so rewarding – but hopefully, nothing happens.
How do you see AI affecting your and your team’s day-to-day?
I think there's maybe different branches that it's assisting in. One is recall – or finding something.
I think in a lot of different ways it's starting to help with maybe mundane tasks that don’t really need a thinker behind them – like does the sheet index match the plans in the set? Are there missing reference details, things like that? So as it gets better at reading and seeing plans, I think we're trying to use it to reduce tasks that aren't the deep-thinking code part of our job, but are still a part of it.
As I understand, you have some newer, younger folks who’ve joined you and are using AI to help expedite their education. If you had to give a gut sense – how much faster or better do you think, or hope, it will be?
As I understand you have some newer younger folks, that have joined you, using AI to help, hopefully, like expedite their education. How, if you had to just like gut-sense, how much faster or better do you think like or do you hope that it will be?
Yeah, so we have a new cohort that we train every year. Train some people who are entry-level – just the ground level – and it's really, I think, mind-blowing to us to see how they're progressing. It's only been a few weeks, but when you start, you don't know what you don't know.
Being able to ask a question and have it recite a bunch of sections you weren’t even aware existed – that's huge. I mean, if you talk to a general person who’s not involved in the code industry, they don’t know what is regulated in a room and how many things that have code requirements for.
So the educational part is really important here. Where it saves me time is by reducing tasks I don't have to do myself. How it's helping them is by helping us train them – it's reducing the time we need to spend on training, and it’s showing them what they’re doing. It can also assist as a co-pilot in helping write comments – and comments are a bit of an art or a style, so they don’t yet have the experience to write concise ones. So it's helping all over the board.
If I were to give a metric – I’m hoping the impact will be exponential. As they go through their careers, we’ll be able to see where they are compared to how we've done this in the past. We've hired and trained people for at least the last fifteen years – every year we try to bring in a new batch of entry-level individuals with varying experiences. Now we’re fifteen years in and can see where they all are. I’m one of those people – trained by our owner, Gein, and started in his class – and I’ll be a benchmark others can look at and compare in terms of years of experience.
One thing I'd love to ask you about is the purpose built nature of Ichi and the trust component. So you're a like you're very detail oriented person. How do you trust these tools?
I would say the biggest thing that surprises me about Ichi is its accuracy. If you had asked me before we used it, I might’ve been one of those people who said, “We’re always going to need a human involved” – especially for things that require interpretation or understanding the intent in order to enforce it. The code is a prescriptive document, and I think the vast majority of what we look at is made up of prescriptive requirements. Having used Ichi – with its brain that’s focused directly on building – the trust that’s built is in how consistently it gets those things right. And it’s not just that I know it’s right – if the person using it doesn’t know, it can cite the code section and send you to the source so you can verify it yourself. So it’s a “trust but verify” kind of thing that builds over time. Having used it for a while now, I’d say its accuracy is the most surprising part for me.
And it’s not just that I know it’s right – if the person using it doesn’t know, it can cite the code section and send you to the source so you can verify it yourself. So it’s a “trust but verify” kind of thing that builds over time. Having used it for a while now, I’d say its accuracy is the most surprising part for me.
I don't know anyone with as much of an encyclopedic memory of codes as you – it's really impressive. But you’ve still said that one of the most interesting things is learning new code sections you didn’t realize were part of a particular context before.
Yeah, I think one of the main things is that codes are constantly changing and growing. New codes come out that you might not be involved in, and there's also more information or evolving interpretations of intent. You might know where a code section is, but for many non-prescriptive issues, that’s just the starting point for understanding where you need to go.
Where I see AI being useful right now is in its ability to grab and compile a bunch of information that wouldn’t be easy to pull together on your own. Take a generator, for example – there are structural, fire, mechanical, and hazardous materials requirements. Being able to bring all that together becomes a learning moment. You come across something you haven’t seen before and realize there was something regulating it that you either weren’t aware of or assumed was always provided. You don’t notice it’s missing until you really look at it – and then you start asking, “Why was that always there?” across all these other submittals.
So yeah, it’s a journey. What we do – like in many professions – is a practice. There’s no moment where you say, “I’ve arrived, I’m a plan checker.” We’re all practicing. We’re getting better, we’re forgetting things – but hopefully, we’re always learning and growing throughout our careers.
Where do you see – if you imagine 5–10 years down the road – how AI will change the industry?
Improvement will come on the design side – less back and forth, and the process won’t be upside down anymore, where the design is done first and then code reviewers get involved. Instead, code will be integrated upfront in the software. Architects and engineers will have code expertise at their fingertips before it reaches the city – and before it becomes a problem.
People in the field, before things are built, will be able to ask questions and get answers – instead of relying on “this is how I’ve done it for 15 years, so I’m just going to keep doing it this way.” It’s hard to predict that far ahead, but I think we’re at the precipice of a big change. And even without AI, just looking at how far the field has come in the last 5 years – it’s already significant.
