<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: TS Lim</title>
    <description>The latest articles on DEV Community by TS Lim (@tslim).</description>
    <link>https://dev.to/tslim</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F429691%2Ffc40b225-d631-4b17-b9de-0c9c91e9cc69.jpeg</url>
      <title>DEV Community: TS Lim</title>
      <link>https://dev.to/tslim</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/tslim"/>
    <language>en</language>
    <item>
      <title>Building AI with AI</title>
      <dc:creator>TS Lim</dc:creator>
      <pubDate>Thu, 19 Nov 2026 04:45:56 +0000</pubDate>
      <link>https://dev.to/tslim/building-ai-with-ai-22m3</link>
      <guid>https://dev.to/tslim/building-ai-with-ai-22m3</guid>
      <description>&lt;p&gt;In my previous posts, I've shared more about my &lt;a href="https://www.clipsoflogic.com/working-with-ai-with-ai/" rel="noopener noreferrer"&gt;AI workflows&lt;/a&gt;, &lt;a href="https://www.clipsoflogic.com/extensions-that-extend-you/" rel="noopener noreferrer"&gt;extensions&lt;/a&gt; and &lt;a href="https://www.clipsoflogic.com/its-just-a-skill/" rel="noopener noreferrer"&gt;tooling&lt;/a&gt; to help manage my administrative tasks at work. I've also explored how you can use &lt;a href="https://www.clipsoflogic.com/learning-anything-with-ai/" rel="noopener noreferrer"&gt;AI to learn anything&lt;/a&gt; by getting it to build interactive guides and learning experiences.&lt;/p&gt;

&lt;p&gt;This time, I want to share some lessons I've learned from building software with AI (coding agents) and also building AI-powered systems at work. Most of these insights come from hands-on experience building a few projects, primarily using &lt;a href="https://pi.dev/?ref=clipsoflogic.com" rel="noopener noreferrer"&gt;Pi.dev&lt;/a&gt; and &lt;a href="https://cursor.com/?ref=clipsoflogic.com" rel="noopener noreferrer"&gt;Cursor IDE&lt;/a&gt;, powered by &lt;a href="https://openai.com/codex/?ref=clipsoflogic.com" rel="noopener noreferrer"&gt;OpenAI Codex&lt;/a&gt; and Cursor subscriptions.&lt;/p&gt;

&lt;h2&gt;
  
  
  Building with AI
&lt;/h2&gt;

&lt;p&gt;First, let's talk about using AI coding agents to design and develop software. I build mostly web applications, so some these lessons may not apply to other types of software. Still, I think many of the principles are general enough to be useful regardless of what you're building.&lt;/p&gt;

&lt;h3&gt;
  
  
  1. Models are getting really good
&lt;/h3&gt;

&lt;p&gt;Every few months, we get bigger and better models from the frontier labs. But things have moved and improved so much that even the smaller, open-weight models are now good enough for coding, especially for well-scoped problems and tasks.&lt;/p&gt;

&lt;p&gt;Frontier models often feels smarter and better because they seem more capable of filling in the gaps when we don't provide enough details. They have stronger reasoning, broader knowledge and a better ability to infer intent, so it can sometimes feel like they are reading your mind. But if we are being honest, most of the time we haven't actually thought through what we want and are just glad they're able to fill in the gaps.&lt;/p&gt;

&lt;p&gt;The danger here is that if we never thought about what we really want, the assumptions made by the models aren't necessarily correct and can take us down the wrong trajectory. Like a really smart person, they can provide really convincing justification for why their proposed solutions make sense. If we don't understand the domain well enough, we can easily be taken along for the ride.&lt;/p&gt;

&lt;h3&gt;
  
  
  2. Plans and research are useful up to a point
&lt;/h3&gt;

&lt;p&gt;Most coding agents have a plan mode, which you should definitely leverage whenever possible. It helps you shape the problem and review the proposed solution before implementation.&lt;/p&gt;

&lt;p&gt;You can also ask them to research alternative solutions and run experiments or spikes to validate assumptions, especially for changes that impact a large portion of your codebase.&lt;/p&gt;

&lt;p&gt;However, remember that models are probabilistic and may not have all the required context, so plans and research can still miss things. The larger the scope you ask an agent to cover, the more opportunities there are for things to go wrong and the harder it becomes for you to comprehend and review the proposed changes.&lt;/p&gt;

&lt;p&gt;I like to ask for diagrams, request flow or other visualizations to help me understand the proposed solution. Your input in scoping the problem, providing constraints and making strategic choices goes a long way towards building the right thing.&lt;/p&gt;

&lt;h3&gt;
  
  
  3. Treat the agent like a fellow developer
&lt;/h3&gt;

&lt;p&gt;This may not be obvious to some of you, but I like to think of and talk to a coding agent as if it were another developer. Brief it as you would someone taking over a project from you and review its work by probing areas you think are important.&lt;/p&gt;

&lt;p&gt;Give it as much context and background information as you can about the problem. Explain what you think the approach should be and discuss the trade-offs. Collaborate with the agent rather than viewing it simply as a machine that spits out code.&lt;/p&gt;

&lt;p&gt;Recently I tried using Nico Bailon's &lt;a href="https://github.com/nicobailon/pi-better?ref=clipsoflogic.com" rel="noopener noreferrer"&gt;pi-better&lt;/a&gt; extension, which rewrites your prompt using &lt;a href="https://felixrieseberg.com/how-to-be-more-ambitious-a-tutorial-for-humans-with-models/?ref=clipsoflogic.com#depth" rel="noopener noreferrer"&gt;Felix Rieseberg's "Depth" prompt&lt;/a&gt;. It grounds the prompt in your existing code and files, giving the agent additional context that you might not have explicitly provided without expanding the scope of the task.&lt;/p&gt;

&lt;p&gt;However, much like working with people, agents can sometimes become overly focused on a particular requirement and over-engineer a solution just to make sure it is satisfied. Even though you're collaborating with the agent, you're still the decision maker. You need to consciously play that role rather than delegate those decisions to it.&lt;/p&gt;

&lt;h3&gt;
  
  
  4. Don't trust the guardrails blindly
&lt;/h3&gt;

&lt;p&gt;One of the best practices is to define an AGENTS.md / CLAUDE.md instruction file containing coding guidelines, guardrails and dos &amp;amp; don'ts for your codebase. These instructions provide the agent with guidance on how it should work and what it should or shouldn't do.&lt;/p&gt;

&lt;p&gt;However, models are non-deterministic. Giving the same prompt twice doesn't guarantee the same result. We shouldn't rely solely on these instructions to ensure compliance. Prefer actual code-based checks such as linters, type checkers and tests, wherever possible. Also, limit the agent's permission and access through your coding harness to prevent it from doing things it shouldn't.&lt;/p&gt;

&lt;p&gt;Beware of &lt;a href="https://en.wikipedia.org/wiki/Reward_hacking?ref=clipsoflogic.com" rel="noopener noreferrer"&gt;reward hacking&lt;/a&gt;, particularly when asking agents to fix failing tests or change existing behavior. They might modify the test to make them pass or change a feature to match what the test expects, rather than addressing the underlying issue. Just because the tests are green doesn't mean the solution is correct.&lt;/p&gt;

&lt;h3&gt;
  
  
  5. Your eyes are still important
&lt;/h3&gt;

&lt;p&gt;In the past, we tended to skip tests or not write enough of them. With coding agents, they will almost always write tests by default, and sometimes it can feel like they write too many. You may need to challenge them on whether all those tests are necessary.&lt;/p&gt;

&lt;p&gt;As agents get better at using computer and browsers, I find them very useful as manual QA testers. You can simply ask them to test the application and explore whether things are working as expected. Since they can also read the related code, the bug reports they produce can be valuable input for the coding agent to investigate and fix.&lt;/p&gt;

&lt;p&gt;That said, nothing beats using the application yourself to discover problems and issues. You have a better understanding of how things should work and it's difficult for an agent to evaluate usability, visual consistency, interaction patterns and overall feel of an application.&lt;/p&gt;

&lt;h2&gt;
  
  
  Building AI-powered applications
&lt;/h2&gt;

&lt;p&gt;Now, let's talk about some of the insights and lessons I've learned building AI-powered workflows and applications. These are based on actual projects I've worked on, but it's obviously not an exhaustive list. The lessons are also limited to the types of features and capabilities I've worked with.&lt;/p&gt;

&lt;h3&gt;
  
  
  1. Look at your RAG extraction and retrieval first
&lt;/h3&gt;

&lt;p&gt;One of the most common types of AI-powered applications I've worked on is a RAG (Retrieval-Augmented Generation) system where you can ask an AI agent to search through a corpus of documents or knowledge base and return relevant information.&lt;/p&gt;

&lt;p&gt;Naturally, the accuracy of the results is critical. It's worth starting at the ingestion stage and checking whether information is being extracted from the source documents properly. You can't retrieve information that wasn't captured in the first place.&lt;/p&gt;

&lt;p&gt;Extracting text is generally more reliable than extracting information from diagrams, graphs and other visual content. Some services can convert these into text descriptions, making the information easier to chunk and retrieve. Don't overlook metadata such as document structure, page numbers and annotations either. These can provide useful context during retrieval and help trace information back to its source.&lt;/p&gt;

&lt;p&gt;Next, review your retrieval strategies and algorithms before tweaking your agent's prompt or switching to a better model. Don't assume your RAG accuracy problem is an LLM problem. First, verify that you have extracted the right information and that your retrieval process is actually finding it.&lt;/p&gt;

&lt;h3&gt;
  
  
  2. You don't always need an LLM
&lt;/h3&gt;

&lt;p&gt;In many agentic workflows, people have a tendency to rely too much on LLMs for things that deterministic code could perform much better and more reliably. LLMs are smart and can handle irregular inputs, but they're also much slower and more expensive than a few simple if statements.&lt;/p&gt;

&lt;p&gt;Here are some examples of things you should generally avoid using LLMs for: calculations, enforcing business rules or permissions, validating data against a fixed set of requirements and orchestrating a predefined sequence of workflows. These are tasks where the expected behavior is known and can be expressed clearly in code.&lt;/p&gt;

&lt;p&gt;Reserve LLMs for situations where you actually need reasoning or where interpreting the input isn't straightforward. For example, understanding a user's intent, extracting information from unstructured content or deciding which workflow should handle a request.&lt;/p&gt;

&lt;p&gt;Even then, you don't always need the most capable reasoning model. System 1 Models or decision models such as &lt;a href="https://typesafe.ai/blog/introducing-system-one-models-and-jev?ref=clipsoflogic.com" rel="noopener noreferrer"&gt;TypeSafe Jev&lt;/a&gt;, &lt;a href="https://developers.openai.com/api/docs/guides/decisions?ref=clipsoflogic.com" rel="noopener noreferrer"&gt;OpenAI Decisions API&lt;/a&gt; and &lt;a href="https://aistudio.google.com/docs/models/gemini-decision-1?ref=clipsoflogic.com" rel="noopener noreferrer"&gt;Gemini Decision 1&lt;/a&gt; allow us to make fast probabilistic decisions within our workflows.&lt;/p&gt;

&lt;p&gt;In fact, you can just ask your coding agent to review your workflow and identify which parts should be handled by deterministic code instead. It can even help you implement them. The key to building a robust agentic workflow is knowing where to use LLMs and just as importantly, where not to.&lt;/p&gt;

&lt;h3&gt;
  
  
  3. Enforce constraints outside the model
&lt;/h3&gt;

&lt;p&gt;Just as you shouldn't blindly trust the guardrails set up around your coding agent, you should enforce and limit a model's capabilities through the harness or runtime it's operating in, rather than relying on prompts.&lt;/p&gt;

&lt;p&gt;If you need to ensure a tool respects the user's access and permission levels, don't rely on the model to pass the correct user information to the tool. The tool itself should enforce those checks and constraints, using trusted identity and permission information that the model cannot manipulate.&lt;/p&gt;

&lt;p&gt;Similarly, you can prompt your agent to always require human approval and implement a Human-in-the-Loop (HITL) approval process but the approval requirements need to be enforced by your harness, tools and runtime. The best way to prevent your model from doing something is to remove its ability to do it in the first place.&lt;/p&gt;

&lt;h3&gt;
  
  
  4. Understand your agentic primitives
&lt;/h3&gt;

&lt;p&gt;It's important to have the right mental model of agents and how they work. By understanding the fundamental building blocks of an agentic system, you'll be able to combine them to create different types of agentic applications and workflows.&lt;/p&gt;

&lt;p&gt;Here are some primitives to get you started. This space is evolving quickly, so make sure you keep up with the latest development.&lt;/p&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Primitive&lt;/th&gt;
&lt;th&gt;What it represents&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;&lt;strong&gt;Model&lt;/strong&gt;&lt;/td&gt;
&lt;td&gt;The underlying intelligence that interprets information, reasons and generates responses.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;&lt;strong&gt;Instructions&lt;/strong&gt;&lt;/td&gt;
&lt;td&gt;What the agent is supposed to do, including its role, objectives and behavioural constraints.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;&lt;strong&gt;Tools&lt;/strong&gt;&lt;/td&gt;
&lt;td&gt;What the agent can do, including interacting with external systems, retrieving information and executing actions.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;&lt;strong&gt;Context&lt;/strong&gt;&lt;/td&gt;
&lt;td&gt;Information available to the model when making decisions, including instructions, conversation history and retrieved information.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;&lt;strong&gt;Session / State&lt;/strong&gt;&lt;/td&gt;
&lt;td&gt;Information maintained during an interaction or task, including progress, intermediate results and execution state.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;&lt;strong&gt;Memory&lt;/strong&gt;&lt;/td&gt;
&lt;td&gt;Information retained for future use, such as user preferences, past experiences and accumulated knowledge.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;&lt;strong&gt;Harness / Runtime&lt;/strong&gt;&lt;/td&gt;
&lt;td&gt;The system that executes and manages the agent, including its environment, tool access and execution lifecycle.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;&lt;strong&gt;Orchestration&lt;/strong&gt;&lt;/td&gt;
&lt;td&gt;How multiple steps, workflows or agents are coordinated to accomplish a task.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;&lt;strong&gt;Guardrails&lt;/strong&gt;&lt;/td&gt;
&lt;td&gt;Controls around inputs, outputs and actions, including permissions, validation and human approval.&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

&lt;p&gt;Use these primitives to build prototypes that emulate agentic products and workflows you see in the market. Design a deep research agent that can plan, conduct web searches across the internet and put together a report. Figure out how to build a Personal Agent like &lt;a href="https://x.ai/bot?ref=clipsoflogic.com" rel="noopener noreferrer"&gt;GrokBot&lt;/a&gt;, &lt;a href="https://instinct.com/?ref=clipsoflogic.com" rel="noopener noreferrer"&gt;Instinct&lt;/a&gt; or &lt;a href="https://muse.ai/?ref=clipsoflogic.com" rel="noopener noreferrer"&gt;Muse&lt;/a&gt;.&lt;/p&gt;

&lt;p&gt;The best way to learn to build AI systems is to build one yourself, with the help of AI, of course.&lt;/p&gt;

&lt;h3&gt;
  
  
  5. Observe, evaluate and optimize your agents
&lt;/h3&gt;

&lt;p&gt;When working with agentic systems, make sure you have some way to observe what's happening behind the scenes. Set up observability tools like &lt;a href="https://mlflow.org/?ref=clipsoflogic.com" rel="noopener noreferrer"&gt;MLflow&lt;/a&gt; to capture traces of your LLMs and tool executions. This allows you to see what exactly is sent to and returned from LLMs, which tools are executed and what results they produced.&lt;/p&gt;

&lt;p&gt;These traces are invaluable when you are trying to figure out what's wrong or how to improve the performance, efficiency and latency of your system. You can even paste the full traces into your coding agent and ask it to help debug the issue. Agents can be quite persistent in trying to complete their tasks and their tool execution trajectories aren't always the most efficient in terms of token usage or number of tool calls.&lt;/p&gt;

&lt;p&gt;You can also setup &lt;a href="https://www.anthropic.com/research/demystifying-evals-for-ai-agents?ref=clipsoflogic.com" rel="noopener noreferrer"&gt;evaluations&lt;/a&gt; (Evals) to regularly check whether your agentic workflows are still behaving as expected as you tweak and update your system. They can help detect model drift and regressions as you improve your prompts, switch models and optimize your workflow.&lt;/p&gt;

&lt;h2&gt;
  
  
  Building in the age of AI
&lt;/h2&gt;

&lt;p&gt;As AI agents continue to improve, they have greatly accelerated our ability to build software and complex systems. Things that used to take weeks can now be done in hours or even minutes. This changes the calculus of what's possible and opens up all sorts of new possibilities for us to explore.&lt;/p&gt;

&lt;p&gt;But I've come to realize that we can only reliably move as fast as our mental models allow us to. When the agent executes faster than our understanding, we lose sight of the bigger picture. We may build something that works, but don't really understand how or why it works. Sometimes, we don't even know if it's what we intended to build.&lt;/p&gt;

&lt;p&gt;Maybe as the models continue to improve and the tooling matures, many of these problems will become a thing of the past. Some of these insights and lessons may no longer be applicable by then. But I suspect that, as builders, we will always need to be clear on what we intend to build.&lt;/p&gt;

</description>
      <category>agents</category>
      <category>ai</category>
      <category>programming</category>
      <category>softwaredevelopment</category>
    </item>
    <item>
      <title>Software is Abstracting the World</title>
      <dc:creator>TS Lim</dc:creator>
      <pubDate>Tue, 14 Jul 2020 07:27:58 +0000</pubDate>
      <link>https://dev.to/tslim/software-is-abstracting-the-world-1ae6</link>
      <guid>https://dev.to/tslim/software-is-abstracting-the-world-1ae6</guid>
      <description>&lt;p&gt;Almost a decade ago, Marc Andreessen penned the famous "&lt;a href="https://a16z.com/2011/08/20/why-software-is-eating-the-world/" rel="noopener noreferrer"&gt;Why Software is Eating the World&lt;/a&gt;" essay which was amazingly prescient. His arguments on why every company needs to be a software company becomes even more evident in our post-COVID-19 reality where companies like &lt;a href="https://www.amazon.com/" rel="noopener noreferrer"&gt;Amazon&lt;/a&gt;, &lt;a href="https://www.ubereats.com/" rel="noopener noreferrer"&gt;UberEats&lt;/a&gt; and &lt;a href="https://www.doordash.com/" rel="noopener noreferrer"&gt;DoorDash&lt;/a&gt; are now considered essential services.&lt;/p&gt;

&lt;p&gt;While I was mapping out the &lt;a href="https://tslim.github.io/concepts/" rel="noopener noreferrer"&gt;concepts around building software&lt;/a&gt;, I noticed a recurring theme across the world of bits and bytes. Almost everything is an abstraction of something else and this I believe, is the reason why software is bound to permeate across every aspect of our lives.&lt;/p&gt;

&lt;p&gt;In simple terms, &lt;a href="https://en.wikipedia.org/wiki/Abstraction_(computer_science)" rel="noopener noreferrer"&gt;abstraction&lt;/a&gt; allows us to ignore certain details to focus on those that we care about. An abstract version of a cat has just enough characteristics (the ears, the whiskers, the body shape etc) for us to still recognize it without being distracted by other less important details.&lt;/p&gt;

&lt;p&gt;Similarly, in the world of software and code, you'll see this pattern repeated over and over again. &lt;a href="https://en.wikipedia.org/wiki/Assembly_language" rel="noopener noreferrer"&gt;Assembly language&lt;/a&gt;, which is an abstraction of the underlying machine code instructions, is then abstracted by &lt;a href="https://en.wikipedia.org/wiki/High-level_programming_language" rel="noopener noreferrer"&gt;higher-level languages&lt;/a&gt; with easier-to-use constructs to encapsulate away the architectural details of the physical microprocessor.&lt;/p&gt;

&lt;p&gt;Libraries, packages, frameworks and &lt;a href="https://en.wikipedia.org/wiki/Application_programming_interface" rel="noopener noreferrer"&gt;APIs&lt;/a&gt; provide useful abstraction for all sort of capabilities in all sort of domains. There's also &lt;a href="https://en.wikipedia.org/wiki/Declarative_programming" rel="noopener noreferrer"&gt;Declarative programming&lt;/a&gt;, a paradigm that shifts a software engineer's focus towards the outcome rather than the steps to achieve it, abstracting the work to the language itself.&lt;/p&gt;

&lt;p&gt;Cloud computing which has abstracted the notion of physical servers using virtualisation and containers, is abstracted further by the rise of &lt;a href="https://en.wikipedia.org/wiki/Infrastructure_as_code" rel="noopener noreferrer"&gt;Infrastructure-as-Code&lt;/a&gt; (IaC) allowing your entire cloud infrastructure be managed and provisioned through configuration files. &lt;a href="https://en.wikipedia.org/wiki/Serverless_computing" rel="noopener noreferrer"&gt;Serverless computing&lt;/a&gt; then took it a step further by abstracting away infrastructure concerns so you can focus only on your code.&lt;/p&gt;

&lt;p&gt;Even application development isn't spared, with the &lt;a href="https://en.wikipedia.org/wiki/No-code_development_platform" rel="noopener noreferrer"&gt;No-code&lt;/a&gt; movement continuing to gain traction. You can leverage these tools to quickly string together a working application without writing a single line of code.&lt;/p&gt;

&lt;p&gt;This isn't just limited to software development by any means. Uber and Lyft turned the act of hailing a cab into an app, abstracting away the complex orchestration between logistics, scheduling and payments. With just a few clicks, you can get someone to mow your lawn, stock up on groceries and even ensure you have a &lt;a href="https://www.esquire.com/style/mens-accessories/g31019735/best-sock-subscription-boxes/" rel="noopener noreferrer"&gt;lifetime supply of socks&lt;/a&gt;.&lt;/p&gt;

&lt;p&gt;Software companies innately understands the power of abstraction. It allows them to hyper focus on their particular problem domain while leveraging existing technologies and services. Internally, cross-functional autonomous teams are assembled to focus on a particular subset of the problem domain, with the aim of unlocking value that's greater than the sum of its parts.&lt;/p&gt;

&lt;p&gt;Once they themselves are successful, they will then become the abstraction layer of that domain for others to build on. With this moat, they then expand to tackle adjacent problems while fortifying their foothold. This is also why we have startups raising millions of dollars trying solve problems that seems too narrow to the uninitiated.&lt;/p&gt;

&lt;p&gt;As consumers, we too reward companies who are able to solve our problems with as little effort from our side, abstracting away all the complexities. Startups in early stages quite often deliver seemingly magical experiences backed by manual work, hoping to eventually codify the process if there's a product-market fit.&lt;/p&gt;

&lt;p&gt;The unintended consequence for turning everything into black boxes is the unfortunate exploitation of the marginalised. Opaque machine learning algorithms turn human labour into variables to be optimized and scrutinized. In fact, in a lot of cases, the existence of a human workforce is merely a placeholder that will eventually be replaced by robots, improving operating margins and efficiency.&lt;/p&gt;

&lt;p&gt;Abstraction is a double-edge sword that if wielded wisely, let us solve difficult problems, by focusing our efforts and leveraging each other. At the same time, we need to be mindful the potential ill effects of ignoring certain details, especially when it impacts our livelihood.&lt;/p&gt;

&lt;p&gt;Software is supposed to be the great leveller, levelling the playing field for everyone. But if we are not careful, it might end up levelling humanity off the field altogether.&lt;/p&gt;

</description>
      <category>startup</category>
      <category>architecture</category>
      <category>development</category>
      <category>learning</category>
    </item>
    <item>
      <title>Concepts around building software</title>
      <dc:creator>TS Lim</dc:creator>
      <pubDate>Sun, 12 Jul 2020 04:16:48 +0000</pubDate>
      <link>https://dev.to/tslim/concepts-around-building-software-e8j</link>
      <guid>https://dev.to/tslim/concepts-around-building-software-e8j</guid>
      <description>&lt;p&gt;Ever since I discovered &lt;a href="https://foambubble.github.io/foam/" rel="noopener noreferrer"&gt;Foam&lt;/a&gt;, I was waiting an opportunity to try it out. It’s a personal knowledge management and sharing system inspired by &lt;a href="https://roamresearch.com/" rel="noopener noreferrer"&gt;Roam Research&lt;/a&gt;, built on &lt;a href="https://code.visualstudio.com/" rel="noopener noreferrer"&gt;Visual Studio Code&lt;/a&gt; and &lt;a href="https://github.com/" rel="noopener noreferrer"&gt;GitHub&lt;/a&gt;. It fits nicely into my development workflow, so that’s a plus.&lt;/p&gt;

&lt;p&gt;Over the weekend, I attempted to build a personal workspace on Foam. Currently I’m using &lt;a href="https://notion.so/" rel="noopener noreferrer"&gt;Notion.so&lt;/a&gt; for my notes, journals and personal information. The Graph Visualisation in Foam (screenshot below) is the main feature I wanted to explore.&lt;/p&gt;

&lt;p&gt;Originally, I attempted to reproduce what I have on Notion but soon realised there really isn’t much links between my information there. Then I decided to document my recent learnings from work around cloud computing, infrastructure and data architecture.&lt;/p&gt;

&lt;h2&gt;
  
  
  Linking Concepts
&lt;/h2&gt;

&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fi%2Ffnivgfx4wavztlqa2iky.png" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fi%2Ffnivgfx4wavztlqa2iky.png" alt="Graph Visualisation for all the concepts" width="800" height="492"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;As I started to link notes between various concepts, I realised there are many more in software development in general. What started from just &lt;a href="https://tslim.github.io/concepts/concepts/cloud-computing" rel="noopener noreferrer"&gt;cloud computing&lt;/a&gt; and &lt;a href="https://tslim.github.io/concepts/concepts/data-architecture" rel="noopener noreferrer"&gt;data architecture&lt;/a&gt;, ends up covering &lt;a href="https://tslim.github.io/concepts/concepts/software-architecture" rel="noopener noreferrer"&gt;software architecture&lt;/a&gt;, &lt;a href="https://tslim.github.io/concepts/concepts/software-development" rel="noopener noreferrer"&gt;development&lt;/a&gt;, &lt;a href="https://tslim.github.io/concepts/concepts/software-engineering" rel="noopener noreferrer"&gt;engineering&lt;/a&gt;, &lt;a href="https://tslim.github.io/concepts/concepts/product-development" rel="noopener noreferrer"&gt;product development&lt;/a&gt; and even &lt;a href="https://tslim.github.io/concepts/concepts/machine-learning" rel="noopener noreferrer"&gt;machine learning&lt;/a&gt;.&lt;/p&gt;

&lt;p&gt;These concepts are by no means the only one that matters. It’s mostly from my own experience and I’m sure there are a lot more out there to be discovered. But this is a good starting point to see how things are inter-related and realised software development is actually quite complicated.&lt;/p&gt;

&lt;h2&gt;
  
  
  Publish and Share
&lt;/h2&gt;

&lt;p&gt;Since Foam uses GitHub as storage, it takes almost no effort to publish the workspace on &lt;a href="https://pages.github.com/" rel="noopener noreferrer"&gt;GitHub Pages&lt;/a&gt;. Combined with the awesome &lt;a href="https://pmarsceill.github.io/just-the-docs/" rel="noopener noreferrer"&gt;Just the Doc&lt;/a&gt; theme, the whole project looks really great &lt;a href="https://tslim.github.io/concepts/" rel="noopener noreferrer"&gt;online&lt;/a&gt;.&lt;/p&gt;

&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fi%2Fjw6s1c72g52xb89fhm7i.jpg" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fi%2Fjw6s1c72g52xb89fhm7i.jpg" alt="Published Foam Workspace on GitHub Pages" width="800" height="586"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;I’ve also started to share this with my team and got lots of great feedback on what concepts I missed and how to use it in general. It can serve as a guide for their career growth in their respective areas.&lt;/p&gt;

&lt;p&gt;If you are a software engineer or interested in some of these topics, check out the site at &lt;a href="https://tslim.github.io/concepts" rel="noopener noreferrer"&gt;https://tslim.github.io/concepts&lt;/a&gt;. You are also welcomed to contribute more notes and concepts at the &lt;a href="https://github.com/tslim/concepts" rel="noopener noreferrer"&gt;repo&lt;/a&gt;.&lt;/p&gt;

</description>
      <category>vscode</category>
      <category>github</category>
      <category>architecture</category>
      <category>showdev</category>
    </item>
  </channel>
</rss>
