Biotech has a developer problem — here’s how to fix it

When I joined Benchling after years building developer platforms at GitHub and Heroku, I expected biotech to feel different. What I didn’t expect was how familiar the underlying problems felt. In most dev-tools companies, developers build tools for other developers. In biotech, they’re trying to build the digital foundation for science itself, but with an ecosystem that hasn’t historically treated developers as first-class citizens.

For decades, wet lab and dry lab work have lived in separate worlds. Scientists run experiments and later developers get handed the aftermath and are asked to “make the data usable.”  

With AI, this pattern is making data problems even more acute. Models can’t operate in real labs unless someone builds pipelines, data contracts, and feedback loops that make the science machine-readable. And that someone is almost always developers, the same people who today are still acting as human APIs, manually stitching together fragmented tools and exporting CSVs like it’s 2008. 

If biotech wants to harness AI, that dynamic has to change. Platforms need to be developer-friendly in order to be AI-ready. Here’s what that means, why it matters, and how we’re building toward that future at Benchling. 

Why “developer-friendly biotech” matters now

In most industries, the shift toward developer-centric platforms happened years ago. Open APIs, reusable integrations, and community-driven ecosystems unlocked massive waves of innovation. Biotech is only beginning that transition, and the stakes are higher.

Here, the data is the IP. Security and governance can’t be optional. But, speed and openness can’t be optional either, not if you want to compete with the pace of modern research or leverage AI in a meaningful way. 

The question the industry has to answer is simple but uncomfortable: 

How do you make it easy for machines, models, and developers to work with biotech data without compromising IP, reproducibility, and security?

At Benchling, this boils down to three core principles we’ve built our platform around: 

• Extensible: An extensible platform is where developers can build custom, interactive experiences directly in the system. For us that means that you can build virtually anything on top of Benchling, or integrate custom logic directly into your Benchling workflows.

• Interoperable: An interoperable platform is where data flows freely between tools, models, and analytics environments. At Benchling, we build our platform so that it connects seamlessly with other tools in your tech stack. 

• Accessible: An accessible platform is one that delivers extra value to your workflows, and in particular also your AI-powered initiatives. For Benchling that means broad coverage across teams and complete documentation of all your data across Benchling products. 

How AI raises the stakes — fast

AI has pushed both the volume and complexity of R&D data to new heights. Teams are now generating and using data at a scale that trains models, powers predictions, and informs decisions in near-real time.

But AI didn’t invent biotech’s data challenges. It simply made them impossible to ignore.

Models need clean data. They need predictable schemas. They need machine-to-machine workflows and APIs that behave the same way at 100 requests as they do at hundreds of thousands. Without that foundation, even the strongest model can't return meaningful scientific insight.

Developers sit squarely at this intersection. They translate scientific intent into data structures that serve both scientists and machines. They architect APIs that move massive datasets securely and reliably. Yet when platforms aren’t extensible, interoperable, or accessible, developers spend more time fighting fragmentation than enabling AI-driven R&D.

But when those conditions are in place?

Biotech gets AI-ready data, faster iteration cycles, and teams where scientists and developers can build together instead of tossing work over the wall. 

So what does a “developer-friendly biotech platform” actually look like?

There are some universal truths from the developer-tools world. Clear, transparent documentation that explains exactly how systems behave. High-quality, predictable APIs that let teams extend the platform confidently. Performance tuned to the scale and complexity of the data.

But biotech layers on something you don’t encounter in the dev-tools world — deep scientific context. 

Developer-friendly biotech means:

• APIs reflect biological concepts

• Terminology maps cleanly between scientists and engineers

• Data structures capture scientific nuance rather than flattening it

• The platform itself teaches developers how the science works

At Benchling, this is at the core of our developer platform. We've built App Canvas so teams can embed custom logic and interactive tools directly into their workflows — 1,900+ apps and counting. Shareable Apps let teams reuse those secure integrations across teams instead of needing to rebuild them. The integration network connects it all to the rest of the stack, connecting with inventory systems, analytics platforms like JMP and Pluto, and data lakes like Snowflake. API v3 ties it together with high coverage and clear usage contracts, so developers know exactly what to expect at scale.

This is scaffolding for a truly open, interoperable biotech ecosystem. And they enable the thing biotech desperately needs more of — scientists and developers co-developing solutions, side by side, from day one. 

The road ahead

Biotech is in the middle of a once-in-a-generation shift. The next wave won’t be defined by tools that are only user-friendly, but by platforms where machines can interpret and act on scientific data as intuitively as humans do.

Organizations that thrive will be the ones that connect their data, tools, automations, and models into a single, adaptive ecosystem. That’s what we’re building toward at Benchling.

My job, and the job of every developer in this industry, is to help biotech cross that bridge from human APIs to real APIs and from fragmented tools to connected systems. This is the moment where biotech starts to feel a little more like the developer ecosystem I came from and where that ecosystem has the chance to reshape how science is done.

Read more about how we’re building the Benchling Developer Platform to support an open, extensible future for biotech.