Thoughtworks Technology Radar Vol. 24 from a bootstrapper's perspective


While waiting for the new edition of Thoughtwork’s Techonology Radar1 – which should arrive any day now – I thought I’d go over the current, volume 24. What can bootstrappers and solo founders take away from the Technology Radar, which is really aimed at enterprises and larger organisations?

Out of the 104 techniques, tools, platforms, frameworks and languages, I have picked three that I found most relevant for bootstrappers and solo-founders.

Techniques: Low Code ⚠️

I have a love-hate relationship with low- and no-code tools. I have extensive experience implementing pretty complex low-code solutions for small enterprises. Their biggest strength is that you can achieve so much so little effort. A trap that many non-developers will fall into when starting their no-code journey: They will suddently try to solve everything with their newly found tools. It’s the typical “When the only tool you have is a hammer, everything starts to look like a nail”-attitude. Developers know this only too well, but we will have (hopefully) gotten better at spotting our bias by now.

The second problem with no-code is the lack of support for solid development practices, like refactoring, versioning, and testing. Thoughtworks agrees:

The problems we see with these platforms typically relate to an inability to apply good engineering practices such as versioning. Testing too is typically really hard.2

What does this mean for bootstrappers? Should I not use low-code / no-code tools at all?

No, not at all. But be keenly aware of their limitations. If you plan to build your whole business on one of these tools, then you will eventually run in one of the aforementioned problems. Someone will make a change in a workflow and suddenly everything stops working. And some tools don’t have any way to revert those changes to the workflow, let alone to the data.

I would suggest limiting the use of no-code / low-code to MVPs, prototypes, and non-critical parts of your business.

Techniques: Simplest possible ML 👍👍👍

Machine Learning (ML) has been all the rage over the last couple of years, with every major cloud provider offering every type of ML model under the sun. Yes, neural networks can do amazing things. But they are expensive and complex – both from financial as well an development/operational perspective.

Thoughtworks suggests to re-think your strategy and not immediately default to using neural networks, when simpler models work fine:

In our experience, teams often choose complex tools because they underestimate the power of simpler tools such as linear regression. Many ML problems don’t require a GPU or neural networks. For that reason we advocate for the simplest possible ML, using simple tools and models and a few hundred lines of Python on the compute platform you have at hand. Only reach for the complex tools when you can demonstrate the need for them. 3

Tools: detect-secrets 👍👍👍

This is another great tool that you should consider including in both your pre-commit-hooks and your CI/CD pipeline.

I am a big fan of automated code-scans, that’s why I have previously suggested using automatic security scans and architecture scans. Yelp’s detect-secrets is another one of those tools. It will try to prevent you from accidentally exposing passwords, keys, and secrets. According to Thoughtworks

Compared to similar offerings, we found that this tool detects more types of secrets with its out-of-the-box configuration.4


See also