AI needs more uncertainty
AI has a confidence problem. Human intelligence was shaped by an uncertain world, and uncertainty is exactly the ingredient AI is missing.
The blog · since 2015
Essays, playbooks and field notes from our CTOs, architects and leads. Written for people shipping real product.
AI has a confidence problem. Human intelligence was shaped by an uncertain world, and uncertainty is exactly the ingredient AI is missing.
AI has a confidence problem. Human intelligence was shaped by an uncertain world, and uncertainty is exactly the ingredient AI is missing.
Agents follow the patterns in the surrounding code more reliably than they follow your rules, so every inconsistent naming scheme and half-migrated architecture is a lesson you're teaching them. The fix isn't another rule file. It's deleting the bad example.
Two agents implemented and reviewed a full architecture refactor across five iterations, with zero lines of production code written by hand. What neither agent caught was four innocent lines in a coordinator that had quietly started making policy decisions.
The Android emulator was never the real cost. The setup was: SDK paths, JDK versions, licence acceptance, a hand-edited config file, repeated on every machine and every new joiner. Claude Desktop takes that off your hands, and the wait that's left is spent reading a plan, not a stack trace.
Codex got the same god class Claude had failed to remove, plus one thing Claude never had: a target architecture. An implementing agent and an independent reviewer then took three rounds to converge, because agents renovate around bad architecture rather than delete it.
A boolean feature flag starts a risky migration everywhere at once. Hashing a stable tenant ID into 100 buckets turns the flag into a dial that only ever widens, and the thresholds matter: percent-of-tenants is not percent-of-work.
Handing an AI your engineering conventions cleans up the code without fixing the architecture. A 5,000-line god class shrank to 500 lines across tidy domain files, and an independent reviewing agent still returned FAIL: authority never left the central object.
Becoming AI-native in three years does not need a three-year technology roadmap. It needs a usage policy, a map of your data, one bounded experiment, and both the enthusiasts and the sceptics in the room. Build the ability to change, not the overhaul.
Infrastructure documentation goes stale because it is a second thing to maintain. Terraform plus cloud discovery plus a few query tools describes the system better than any diagram, and leaves humans to document only the why behind the odd decisions.
Hitting a wall with AI usually means hitting the wall of the chat box, not the wall of AI itself. Give a model memory, tool access and a browser, and the interfaces we built for humans become the integration. The bottleneck stops being the model and starts being your imagination.
Sentry silently drops transactions that blow past its span and payload limits, so the heaviest background jobs are exactly the ones missing from your dashboard. Capping span count and truncating span descriptions on bytes brings them back.
An AI-assisted note system captures faster than you can understand, and the gap compounds into internalisation debt: a dense graph of links attached to a thin mental model. The test of a knowledge base is not how much it holds, but how much you could still explain with the file closed.
A practical guide to evaluating fractional CTO providers in Belgium: the three criteria that matter most, how the four provider types differ, what an engagement should cost, and when a full-time hire is the better call.
Teams accrue learning debt when shipping always beats getting better. It stays invisible on every dashboard, compounds fastest under AI-speed change, and quietly caps the ceiling a buyer is willing to price.
Synthetic checks like curl and Lighthouse measure a request your users never make. A HAR file captures the real logged-in flow, and handed to Claude alongside the codebase it turns a wall of JSON into a ranked list of fixes that get implemented in the same conversation.
AI & agentic tooling
We use AI and agentic workflows daily. From coding agents to AI strategy, we help teams adopt what actually works.
Read our AI insightsEngineering leadership
We <3 sharing content for CTOs, senior engineers and founders with real-world insights.
40+ expert authorsFor investors & VCs
Technical audits, codebase assessments, and risk analysis for portfolio managers making investment decisions.
Explore our investor contentOur staff engineers will get your software (back) on track and lead your engineering team. We prepare our exit from day one: our goal is to bring the software and the team to the right level for us to get out again.
Learn more about software engineeringOur CTOs in residence (or fractional CTOs) inject technical leadership into your team, ensuring the technical aspect of your business gets exactly the right attention.
Learn more about CTO as a serviceWe're the market leader for technical due diligence of SaaS companies in Belgium and The Netherlands, servicing clients all over the world. Investors appreciate our smooth process, sharp opinions and top-notch reports.
Learn more about auditingBehind-the-scenes support for your CTO or technical lead to enhance their skills, drive effective change, and grow their teams confidently.
Learn more about CTO coachingWe can help you find the best talent to join your team. Unlike other recruiters, we are CTOs and software engineers ourselves, and through getting our hands dirty, we know the skills required to make an impact.
Learn more about recruitingSometimes your struggles go beyond the technical. Do you need help evaluating the roadmap or sharpening the vision? What should you build in order to scale? Our product management experts know the way.
Learn more about product management