Being Technical Writer
    • Home
    • Categories
    • Videos
    AI

    I Built a Gamified 365-Day AI Learning App in One Evening—Want to Join Me on This Journey?

    <a href='/aboutme/'>Gaurav Trivedi</a> Gaurav Trivedi
    Oct 22, 2025
    18 min read
    I Built a Gamified 365-Day AI Learning App in One Evening—Want to Join Me on This Journey?

    Day 49. November 8, 2025.

    I’m looking at my progress dashboard right now: 590 XP earned. Current streak: 49 consecutive days. I’ve built 6 working Make.com automations that I actually use daily. I’m halfway through Season 2, deep in ServiceNow territory.

    I’m not an expert yet. But I’m not the person I was 7 weeks ago either.

    Here’s what nobody tells you about transformation: you can’t see it happening. You only see it when you look back.

    Rewind to Day 0. September 20, 2025.

    I was doing what I do best: productive procrastination. Tea cooling on my desk, seventeen browser tabs open with “Top 10 AI Courses to Take in 2025,” and that familiar sensation of drowning in options while making zero progress.

    Here’s the thing nobody wants to admit: we’re not afraid of failure. We’re afraid of wasting our effort on the wrong thing. What if I spend 6 months learning TensorFlow and the market wants LangChain? What if I master Make.com but employers want ServiceNow? What if AI makes all this obsolete before I even finish?

    So I built something for myself. Not to sell. Not to launch a course. Not to become a guru.

    I built it because I was tired of waiting for permission to start.

    That was 49 days ago. And now I’m sharing this because maybe you’re in the same place I was 7 weeks ago.

    The conversation that wouldn’t leave my head

    Two weeks before I built this app, I had coffee with my former colleague—let’s call him Raj. He’s been in tech for 15 years. Solid engineer. Good problem solver.

    He looked exhausted.

    Me: “How’s work?”

    Raj: “I just spent three weeks building a data pipeline. Took me 80 hours. Then I showed my manager what Cursor could build in 45 minutes. He just stared at me and said, ‘So… what do we need you for?’”

    Me: “That’s brutal. What did you say?”

    Raj: “I didn’t know what to say. I’ve been Googling ‘AI-proof careers’ every night since.”

    This conversation haunted me because Raj isn’t lazy or incompetent. He’s just caught in the trap most of us are in: we’re good at execution, but AI is getting better at execution faster than we’re getting better at anything else.

    Here’s the mind-bending part: learning how to use AI tools won’t save you. Because by the time you master Claude or ChatGPT, there’s a better model. By the time you learn prompt engineering, AI doesn’t need prompts anymore.

    So what makes someone irreplaceable?

    I spent two weeks obsessing over this question. And I realized: the people who won’t be replaced aren’t the ones using AI—they’re the ones designing the systems that use AI.

    Not “I can prompt an LLM.” That’s table stakes.

    But “I can architect an enterprise workflow that routes customer inquiries through AI triage, escalates to humans based on confidence scores, learns from human corrections, and integrates with legacy systems while maintaining governance and audit trails.”

    That’s a different game entirely.

    That’s what I needed to learn. Not tools. Not models. Systems thinking for an AI-native world.

    So I built a learning path for myself. Then I turned it into an app. Because if I’m going to spend 730 hours on something, I want to see the XP bar fill up.

    What My 365-Day AI Architect App Actually Provides

    Feature What It Solves How It Works Why It Matters
    365-Day Structured Path Decision fatigue & scattered resources One clear task per day, max 2 hours No more "what should I learn today?"
    4 Gamified Seasons Invisible progress & motivation XP, streaks, badges, milestones You see yourself leveling up
    Warm-up Wins Procrastination & overwhelm 5-minute starter task each day Breaks the "too hard to start" barrier
    No-Code → AI Pipeline Imposter syndrome & skill gaps Make.com → ServiceNow → LangChain Builds confidence progressively
    Certification Checkpoints Lack of tangible proof Real certs: ServiceNow CSA, AWS, etc. Resume-worthy credentials

    The pattern here isn’t revolutionary tech—it’s behavioral design. Duolingo proved gamification works for language learning. GitHub proved streaks drive consistency. I’m applying the same psychology to becoming an AI Solutions Architect.

    How I built this in one evening (confession: I almost didn’t)

    Friday, 7 PM. I opened VS Code with grand plans.

    Friday, 7:03 PM. I was reading Next.js documentation because “modern apps need React, right?”

    Friday, 7:47 PM. I was comparing PostgreSQL vs. MongoDB because “scalability matters, right?”

    Friday, 8:15 PM. I had twelve tabs open about Docker Compose.

    Then I caught myself. I was doing the exact thing I built this app to stop doing: optimizing for a future that doesn’t exist yet.

    I deleted everything and asked myself: “What’s the absolute minimum stack that lets me mark a task as done and see a streak counter?”

    Answer: Flask. SQLite. Plain HTML.

    That’s it.

    The entire stack (built in ~4 hours)

    Backend: Flask 3 (~200 lines total—I can read the whole app in one sitting)

    Database: SQLite (it's just a file—delete it to reset, back it up with Dropbox)

    Frontend: Jinja templates + Bootstrap via CDN (zero build step, zero npm)

    Deployment: Runs anywhere Python runs

    ✓ Three commands from clone to running: venv → pip install → python app.py

    Is this the “best” tech stack? Absolutely not.

    Will it scale to 10,000 users? Nope.

    Will it look impressive on Hacker News? Doubtful.

    But was it running on my machine the next morning when I needed to mark Day 1 complete? Yes.

    And that’s the only metric that mattered for a tool I built for myself.

    You’re welcome to use it too. Fork it. Modify it. Rewrite it in Rust if that makes you happy. The goal isn’t the perfect app—it’s the completed journey.

    The 365-day journey: Four seasons of transformation

    The syllabus isn’t random. It’s structured like an RPG campaign with four distinct “seasons,” each building on the last:

    Your 12-month transformation arc

    Season 1: The Connector (Months 1-3)
    • ✅ Master Make.com automation
    • ✅ Build Airtable workflows
    • ✅ Integrate APIs without code
    • ✅ First capstone: Gmail→Slack→Drive pipeline
    Season 2: The Builder (Months 4-6)
    • ✅ ServiceNow CSA Certification
    • ✅ Enterprise workflow design
    • ✅ Python scripting basics
    • ✅ Capstone: IT ticketing automation system
    Season 3: The Architect (Months 7-9)
    • ✅ LangChain & vector databases
    • ✅ Responsible AI frameworks
    • ✅ Decision dashboards (Tableau/PowerBI)
    • ✅ Capstone: Context-aware chatbot
    Season 4: The Governor (Months 10-12)
    • ✅ AI governance & ethics
    • ✅ Human-in-the-loop systems
    • ✅ Portfolio-worthy final project
    • ✅ Capstone: Full AI workflow you can demo

    Notice the progression? You’re not diving into LangChain on Day 1. You start with no-code tools that give you wins immediately, build enterprise credibility with ServiceNow, then graduate to AI architecture.

    By Month 12, you’re not just someone who “knows AI”—you’re someone who can design, build, deploy, and govern AI systems in real organizations.

    “Most courses teach tools. This teaches a career path. There’s a difference.”

    The deliberate pacing: Each day caps at 2 hours. This isn’t a bootcamp that assumes you’re unemployed and living on ramen. It’s designed for people with jobs, families, and lives who want to transition without burning out.

    The gamification that actually works

    Here’s where I stole shamelessly from every addictive app you’ve ever used:

    Dopamine hits built into every day

    XP Points Each task worth 10+ XP, warmup bonus adds +2—you level up visibly
    Streaks GitHub-style consecutive day tracking—seeing that number grow is weirdly motivating
    Badges Unlock "Level 1—Momentum" at 100 XP, "Deep Habit" at 21-day streak, etc.
    Warm-ups 5-minute starter task to beat procrastination—"Just open Make.com" counts
    Notes Field Track reflections, aha moments, and questions—your future self will thank you
    Progress Bar Visual completion percentage—watching it fill from 0% to 100% over 12 months

    Here’s something weird I discovered about myself:

    I’ll skip a workout. I’ll postpone writing. I’ll delay important work.

    But I won’t break a GitHub contributions streak. Something about seeing that green grid with one gray square is psychologically intolerable.

    So I weaponized that quirk against my own procrastination.

    The “warm-up wins” are my favorite hack. Every day starts with a ridiculously easy task: “Open Make.com.” That’s it. Just open it.

    Sounds stupid, right?

    But here’s the thing: starting is a phase transition. Going from “not doing the thing” to “doing the thing” requires massive energy. Once you’ve opened Make.com, you’re 80% of the way to actually building a scenario. The activation energy is spent.

    I proved this to myself on Day 23. I was exhausted. Didn’t want to learn anything. But the streak was at 22 days. So I opened ServiceNow “just to preserve the streak.”

    Thirty seconds later, I was building a workflow. Two hours later, I’d completed the day’s task and taken notes.

    Would I have done that without the streak pressure? Absolutely not.

    Is this manipulating myself with gamification psychology? Yes.

    Does it work? Embarrassingly well.

    The XP, badges, progress bars—they’re not just decoration. They’re making the invisible visible. You’re not vaguely “learning AI.” You’re at 247 XP with a 14-day streak and “Level 2—Flow” badge. That’s concrete. That’s a different identity. That’s proof.

    What a typical day actually looks like

    Let me show you what “2 hours of learning” actually means in practice. Here’s Day 47 (middle of Season 2):

    Day 47: ServiceNow—Building Your First Incident Workflow

    • Warm-up (5 min): Log into your ServiceNow developer instance. Click around. That’s it.
    • Main Task (90 min):
      • Watch the 30-minute “Incident Management” tutorial
      • Build a basic incident creation form
      • Configure assignment rules (when ticket created → assign to support queue)
      • Test with 3 sample incidents
    • Reflection (10 min): Note in app: “Assignment rules are basically IF/THEN with extra steps. Reminds me of Make.com logic.”
    • XP Earned: 10 + 2 (warmup bonus) = 12 XP
    • Certification Progress: 47% toward ServiceNow CSA exam

    Notice what’s not here:

    • No vague “learn ServiceNow”—you’re building a specific thing
    • No 6-hour rabbit hole—the task is scoped to what you can finish
    • No floating knowledge—it’s building toward the CSA certification
    • No isolation—you’re connecting it to what you already know (Make.com logic)

    What this won’t do (the part most people skip)

    I need to be honest about what you’re signing up for.

    This is not a magic pill. It’s 730 hours of showing up when you don’t feel like it. It’s learning ServiceNow forms on a Tuesday night when your friends are watching football. It’s debugging Python on a Saturday morning when you’d rather sleep in.

    Let me tell you what happened on Day 47.

    I was supposed to build my first incident workflow in ServiceNow. But my son had a school project due. One of my colleagues scheduled an unexpected call. My wife asked if I could handle dinner because she had a deadline.

    By 9 PM, I was exhausted. The streak was at 47 days. All I had to do was skip one day. Start fresh tomorrow.

    But here’s the thing about streaks: they work because they guilt-trip you. And guilt is an underrated motivator.

    So I did the warm-up. Just logged into ServiceNow. Told myself I’d preserve the streak with the bare minimum.

    Three hours later (yes, I went past the 2-hour cap), I’d built the workflow, broken it twice, fixed it, and actually understood assignment rules.

    Was it sustainable? Probably not.

    Did it work? Yes.

    Here’s what I can’t solve for you:

    Finding 730 hours

    This assumes 2 hours/day for a year. That's watching one less Netflix episode daily. Scrolling Instagram less. Waking up earlier. Only you know if you can trade that.

    ~$500-800 in certifications

    The app is free. The syllabus is free. But ServiceNow CSA prep, AWS courses, and cert exams cost money. You can skip them, but you'll miss the credentials.

    Solo accountability

    No cohort. No Slack community (yet). No coach. Just you, the progress bar, and your own willpower. If you need external accountability, find an accountability partner.

    No job guarantee

    This gives you skills and proof. It doesn't guarantee a job offer. You still need to network, apply, interview, and convince someone to bet on you. The app can't do that part.

    I built this for myself because I needed structure, not because I had answers. I’m sharing it because maybe you need the same structure. But I’m not pretending it’s easy or guaranteed.

    Why I’m telling you this (and what I’m asking)

    Here’s the truth: I built this for myself. Not to sell courses. Not to build a SaaS. Not to become a LinkedIn influencer.

    I built it because I was tired of feeling left behind while everyone else seemed to “get” AI and I was still stuck in tutorial hell.

    But then I realized something while building it: accountability works better when it’s public.

    If I tell myself “I’m going to learn AI this year,” I’ll quit by February.

    If I tell the internet “I’m documenting 365 days of learning with receipts,” I’m way more likely to follow through. Pride is a powerful motivator.

    So here’s what I’m doing:

    I’m on Day 49 right now. 316 days to go.

    I’m posting weekly updates. Real progress, real struggles, real XP counts. No highlight reel, no guru posturing. Just “here’s what I learned this week and here’s where I got stuck.”

    When I hit Day 100, I’ll post a detailed breakdown: what worked, what didn’t, the exact certifications I’m pursuing, the projects I’ve built, and whether this approach actually delivers on the promise.

    You can do one of three things with this:

    1. Fork the repo and join me. Start your own 365-day journey. We can compare notes, share breakthroughs, commiserate over ServiceNow documentation together.

    2. Follow along without committing. Browse the syllabus. Read my weekly updates. See if this approach resonates before diving in.

    3. Ignore this entirely. Maybe you already have a system. Maybe this isn’t your style. That’s fine. I built this for me, and I’m sharing it in case it helps you. No pressure.

    But if you do join, here’s what I want from you:

    When you hit Day 50, DM me your progress. Tell me what’s working, what’s broken, what you’ve built. I want to know if this system works for more than just me.

    Tag your journey with #365ToAIArchitect. Share your wins, your failures, your “I can’t believe I spent 3 hours debugging this” moments. Let’s make this less lonely.

    And if I disappear after Day 100? Call me out. Seriously. Accountability works both ways.

    I don’t know if I’ll make it to Day 365. But I know 49 days ago, I didn’t think I’d make it to Day 50. And here I am.

    How to actually start (three paths, pick one)

    Path 1: Run It on Your Machine (5 minutes)

    git clone https://github.com/gautriv/ai-solution-architect
    cd ai-solution-architect
    python3 -m venv venv
    source venv/bin/activate  # On Windows: venv\Scripts\activate
    pip install -r requirements.txt
    python app.py
    

    Open http://127.0.0.1:5000/ and you’ll see Day 1 waiting for you.

    That’s it. No Docker. No config files. No “wait, what version of Node do I need?”

    Path 2: Deploy It Somewhere (10 minutes)
    Fork the repo, push to your GitHub, connect to Render or Railway or wherever. Now you can check your progress from your phone while pretending to pay attention in meetings.

    Path 3: Browse Without Committing
    Not ready to start but curious? The entire syllabus is in data/syllabus.json. Clone it, read through 365 days of curriculum, decide if this is for you.

    No pressure. No email signup. No “limited-time offer.” The repo will be there when you’re ready.

    What happens on Day 1?

    You create a Make.com account. You watch a 5-minute getting started video. You build your first automation (probably Gmail to Slack or something equally simple).

    No Python. No cloud credentials. No imposter syndrome about “real developers.”

    Just one small win. Then another. Then another.

    By Day 90, those small wins compound into something you can actually put on a resume.

    The question you’re really asking (and the one you should be)

    You’re probably asking: “Can I really do this? Can someone with my background learn AI architecture in 365 days?”

    Wrong question.

    The right question is: “What happens if I don’t?”

    Let me paint you two futures.

    Future A: You Start

    It’s September 2026. You’ve got 3,650 XP. ServiceNow CSA certificate. A GitHub repo with real projects. You’re not an expert, but you’re not a beginner either. You can architect an AI workflow from scratch. You understand vector databases, not just in theory, but because you’ve debugged them at 11 PM on a Tuesday.

    When someone asks “Can you build an AI system that handles customer support tickets with human escalation?” you don’t freeze. You say “Yes, and here’s how I’d approach it.”

    You’re not competing with AI. You’re designing the systems that use it.

    Future B: You Don’t Start

    It’s September 2026. You’re still reading “Top 10 AI Trends” articles. Still planning to “learn AI properly next quarter.” Still watching other people land roles you convinced yourself you weren’t ready for.

    The gap between you and the people who started hasn’t shrunk. It’s grown. Every day they were building, you were planning to build.

    And here’s the really uncomfortable part: both futures take the same amount of time to arrive.

    316 days from now will come whether you start today or not. The only difference is what you’ll have to show for it.

    Here’s what I know about myself:

    I’ve had moments where I wondered if this was pointless. I’ve wanted to quit. There have been days where emergencies came up and I was tempted to skip “just this once.”

    But I also know that on Day 365, I want to look back and see proof. Not perfect execution—proof of persistence.

    Milestones I’m tracking:

    • Day 49 (today): 6 working Make.com automations I use daily, 590 XP, 49-day streak, Season 2 underway
    • Day 100 (51 days away): Detailed progress report—what’s working, what’s not
    • Day 180 (131 days away): ServiceNow CSA certified (that’s the goal)
    • Day 270 (221 days away): Production LangChain app solving a real problem
    • Day 365 (316 days away): Portfolio and confidence to interview for AI architect roles

    If you’re reading this and feeling that mix of “I want this” and “but what if I fail?”—that’s the right feeling.

    The app is ready. The path is clear. The first task takes 5 minutes.

    The only question is: which September 2026 do you want?


    Repository: github.com/gautriv/ai-solution-architect
    Current Status: Day 49/365 (November 8, 2025)
    Next Major Update: Day 100 detailed breakdown
    Hashtag: #365ToAIArchitect
    Questions/Want to Connect? Drop a comment or find me on [Twitter/LinkedIn/wherever you are]

    Let’s see where we both are in 316 days. I’ll be here with receipts—the good, the bad, and the “I can’t believe I thought I could learn this in 2 hours” moments.

    See you at Day 100.

    Previous Article

    I Tested Google I/O 2025 Tools Over the Weekend—Here's What Actually Changes for Technical Writers

    A weekend hands-on test of Google I/O 2025 AI tools from a technical writer's...

    Share This Post

    You Might Also Like

    I Tested Google I/O 2025 Tools Over the Weekend—Here's What Actually Changes for Technical Writers
    I Tested Google I/O 2025 Tools Over the Weekend—Here's What Actually Changes for Technical Writers
    May 25, 2025
    From prototype to co-pilot - Launching the mini AI writing assistant for technical style enforcement
    From prototype to co-pilot - Launching the mini AI writing assistant for technical style enforcement
    May 15, 2025
    Building an AI Technical Writing Assistant - Transforming Documentation with Machine Learning
    Building an AI Technical Writing Assistant - Transforming Documentation with Machine Learning
    Apr 11, 2025
    View All Posts

    Categories

    Jekyll Explore articles
    Technical Writing Explore articles
    Git Explore articles
    Innovation Explore articles
    AI Explore articles

    Join the Discussion

    Being Technical Writer

    Empowering technical writers with insights, innovations, and industry best practices to excel in the ever-evolving world of technical communication.

    Quick Links

    About me Contact

    Resources

    Site analytics

    Premium Courses

    Contact me for one-on-one mentoring sessions.

    Like the content?

    © 2025 Being Technical Writer. The opinions expressed in this blog are my own, and not those of my employer.

    Privacy policy