Commonplace note is a personal compilation of knowledge - quotes, stories, and my observations on artilces I read on topics that interest me. These range from wealth, learning, networking, life, spirituality, homeschooling, and more.
I try to explore an idea out in public (in line with my learning framework), sometimes even without agreeing to it. Think of these as scrap notes.
Habits that have a high rate of return in life:
- sleeping 8+ hours each day
- lifting weights 3x week
- going for a walk each day
- saving at least 10 percent of your income
- reading every day
- drinking more water and less of everything else
- leaving your phone in another room while you work
time for introspection and an analysis of the most common errors I’ve observed in architectural practices
As I read through this fantastic article, I kept nodding at each mistake because I could recognize the time I made each of them.
Being a “Seagull Architect”
Seagull Architects are highly concept-oriented but disconnected from real-world issues and complexity.
Acting Like a “Factory Worker Architect”
Some architects fall into the trap of behaving like factory workers, mechanically churning out designs, diagrams, or decisions without truly engaging with the bigger picture.
Falling Into the “Golden Hammer Syndrome”
Falling Into the “Golden Hammer Syndrome”. This issue is often exacerbated by company fossilization — when architects stay in the same company for too long without exposure to external ideas or trends. Over time, they become overly comfortable with the tools and processes they know, reluctant to explore new possibilities.
As I age in this industry, I often wonder if I'm becoming a fossil. I keep myself updated by continuous learning.
Succumbing to “Résumé-Driven Development (RDD)”
when decisions are driven more by what looks impressive on the architect’s résumé than by what is actually needed for the project. Architects guilty of RDD prioritize trendy tools, frameworks, or methodologies to boost their personal career prospects, often at the expense of practicality and project success.
I can stand upright and say I have not committed this one fatal mistake. I encourage my team to choose "boring" technologies so that we don't have to be woken up at odd hours with escalations.
Trying to be “Nostradamus”
While strategic thinking is essential, attempting to foresee and address every potential issue far in advance often leads to a lack of agility and results in rigid systems that fail to adapt to changing requirements or technologies.
Pre-mature optimization is a curse which sometimes we can't avoid, especially sitting in IT services organizations. When client demand a "google" type scaling for their startup, you can only argue to some extent. When they threaten to take the business to another org, you have to shut-up and do what the client asks.
Being trapped by the Vendor’s Siren Song
overly trusting and receptive to vendors and commercials, often in the name of "better support" or "guaranteed security." This naivety can lead to costly decisions that prioritize vendor solutions over more practical or open-source alternatives.
I see this mistake so much in 2025 with the peak GenAI siren song. Everyone goes by what the vendors say in their blogs and using their tech only to get burnt as the project goes on. When there is a tight deadline there is a temptation to go with what we read in the docs. That is the time to go with "boring" tech choices.
Lacking Competence in Non-Functional Requirements (NFR) Collection and Integration
A very important skill for Solution Architects is their ability to competently collect and account for Non-Functional Requirements (NFRs). These requirements, which define the system’s qualities rather than its functions, are critical for creating a solution that is not only functional but also robust, scalable, and adaptable to organizational needs.
NFRs are mostly not discussed but come to bite at the wrong time at the wrong places. A good architect should always lead the discussion towards NFRs and capture the details as much as possible. If a system doesn't meet NFRs it is squarely the failure of the architect in the project.
The opposite of dystopia isn’t utopia—which doesn’t exist. It’s “protopia”: a world that is always getting better, but is never perfect. Such a world always has new problems to solve, including some problems created by the old solutions. There is plenty of conflict and intrigue in such a world, and plenty of room for heroes and villains.
The Prime Minister is throwing the full weight of Whitehall behind this industry by agreeing to take forward all 50 recommendations set out by Matt Clifford in his game-changing AI Opportunities Action Plan.
Their action plan is made of three sections:
Invest in the foundations of AI: We need world-class computing and data infrastructure, access to talent and regulation
Push hard on cross-economy AI adoption: The public sector should rapidly pilot and scale AI products and services and encourage the private sector to do the same. This will drive better experiences and outcomes for citizens and boost productivity
Position the UK to be an AI maker, not an AI taker: As the technology becomes more powerful, we should be the best state partner to those building frontier AI. The UK should aim to have true national champions at critical layers of the AI stack so that the UK benefits economically from AI advancement and has influence on future AI’s values, safety and governance
What is the objective?
Our ambition is to shape the AI revolution on principles of shared economic prosperity, improved public services and increased personal opportunities
All political and government proposals sounds exciting. We will see how it translates on the ground. Even if they achieve 50% of what they plan to do, it will kickstart growth in their economy.
AI is useful for a real-world task only if the cost for the AI to do the thing and for you to check its work is less than the existing solution.
Milan provides few good and bad examples.
Though it was written in Oct 2020, it seems to be true even now.
As a Hackernews comment said,
don't use it for something that you aren't able to verify or validate
and
the lower quality output comes much faster than you can generate
I've been using GenAI tools for designing this blog, coding an app, editing blog posts, and writing policy documents for IT organization. In each of these cases, GenAI tools (Windsurf or Open WebUI) generates 60 - 70 percent of the output quickly, which I can validate and verify quickly. Then I start to fill-in the rest.
In this process, I've discovered that even though the GenAI models are probabilistic their output is deterministic [1] since they can be validated.