Every year since I started writing and working in tech, I’ve written a yearly recap — not just to list what I did, but to pause, acknowledge growth, and remind myself why I build in the first place.
2025 was a meaningful year for me. A year of graduating, publishing my add-ons to the Workspace Marketplace, teaching developers, experimenting with AI agents, and contributing to the Google Workspace and Gemini ecosystem in every way I could.
A Personal Milestone: Graduating as an Engineer
This year, I completed my Bachelor of Technology in Computer Engineering from NMIMS, Mumbai. It’s hard to overstate how special that moment was. Balancing final-year projects and exams with consistent writing, building tools, recording tutorials, and community engagement was challenging at times — but it all reinforced the discipline and problem-solving mindset that engineering instilled in me.
What starts as small experiments or hobby projects can grow into meaningful work when done consistently.
Graduating marked the close of one chapter and the exciting start of another, where I can now focus even more deeply on creating practical AI and automation solutions.
I graduated :) - Engineering was fun while it lasted. | Aryan Irani | 22 comments
I graduated :) - Engineering was fun while it lasted. | 22 comments on LinkedIn
linkedin.com
Mumbai Tech Week — Sharing My Work on a Big Stage
One of the year’s highlights was presenting at Mumbai Tech Week 2025, thanks to GDG MAD Mumbai and Faiz Malkani. I demoed two Gemini-powered projects live:
- Turning Google Docs into AI-powered podcasts using Google Cloud
- Generating speaker notes in Google Slides with Gemini 2.0 Flash and Apps Script
A Different Kind of Stage — My TEDx Talk
2025 also brought a deeply personal speaking experience — my TEDx talk at TEDxGEA Youth. This was the first time I stood in front of a crowd not to dive into tech tutorials or code demos, but to open up about myself, my journey, and the messy, human side of building a career in tech.
At 21, I shared how I went from a directionless 15-year-old to a technical writer and co-founder of MadLads Studios. It wasn’t about expertise or polished skills — it was about curiosity sparking everything: starting with simple book reviews, automating Gmail tasks, and embracing uncertainty along the way.
I talked about the messy middle of rejections, learning through failures, the importance of consistency over perfection, saying no to avoid burnout, finding joy in the process (my take on “JOMO” — Joy of Missing Out), and how belief from others (like a professor who trusted me to teach Python) ignited my path before I believed in myself.
Stepping outside my tech comfort zone felt vulnerable, but it was rewarding to connect on a more human level. If you’re navigating your own journey, I hope it resonates. Watch it here: Aryan Irani | TEDxGEA Youth (Uploaded October 2025).
Product Launches: My First Google Workspace Add-ons
One of the biggest professional steps this year was shipping two add-ons to the Google Workspace Marketplace — my first publicly available products built for real users:
SmartCheckbox — A no-code tool that automates checkbox updates in Google Sheets based on custom rules. Ideal for task management, approval workflows, and dynamic trackers. It includes real-time refresh, audit logs, and an intuitive sidebar setup. Install it here.
Sheet Protector Pro — One-click sheet protection with smart features: protect formulas/ranges, easy undo, audit history, and controlled editor sharing. Built for teams who need secure collaboration without the hassle. Install it here.
Launching these taught me invaluable lessons about user experience, Marketplace guidelines, permissions, and iterative polishing. Seeing people install and use them has been incredibly rewarding.
Live Trainings: Getting Started with Google Apps Script
I also ran my comprehensive Mastering Google Apps Script live training program — taking participants from beginner concepts to building advanced automations across Google Workspace. Watching attendees gain confidence, script their own solutions, and apply them to real workflows was one of the most fulfilling parts of the year.
The program is still available for anyone looking to level up their Workspace automation skills: Mastering Google Apps Script — Aryan Irani.
Gemini for Google Workspace Fundamentals — O’Reilly
I’m also thrilled to share that I’ll be leading an upcoming Gemini for Google Workspace Fundamentals live training with O’Reilly on January 22, 2026. It’s a hands-on course covering how to leverage Gemini’s AI capabilities across Docs, Sheets, Gmail, and more to boost productivity. From writing effective prompts to automating workflows, I’m excited to help developers and professionals unlock Gemini’s potential. Learn more and register here.
Writing: A Strong Year of Deep Technical Content
My writing this year became more focused on production-ready solutions — especially building AI agents that don’t just respond, but take meaningful actions. I published consistently on Medium (largely through Google Cloud Community), with a clear progression: starting with core Workspace + Gemini automations, moving into Agent Development Kit (ADK) basics, then deep into dynamic agents with MCP Toolbox for Databases, and finally deployment and Workspace integration.
Google Workspace & Gemini Automations
- Fetch BigQuery Data to Google Sheets using Google Apps Script (Feb)
- Generate Speaker Notes in Google Slides using Gemini 2.0 Flash and Google Apps Script (Feb)
- Turn Google Docs into AI-Powered Podcasts with Google Cloud (Feb)
- How to Use Service Accounts with Google Apps Script (Mar)
- Turn Google Docs Comments into Actionable Summaries with Gemini and Apps Script (Oct)
- Automating Google Drive Labels with Google Apps Script: A Step-by-Step Guide (Dec)
- Automating Google Drive Labels with Gemini and Apps Script (Dec)
Getting Started with ADK & Practical Agents
- Your First ADK Agent: Building a Google Tasks To-Do Manager (Jun)
- Build Your Own AI Google Calendar Assistant with Agent Development Kit (Sep)
- MCP Toolbox for Databases Course Series (Connecting Agents to Live Data)
- Build an AI That Connects to Your Database: MCP Toolbox for Databases Course (Jul)
- The Data Gap | Connecting an Agent to a LIVE Database (Jul)
- Build Your First AI Agent with ADK (Jul)
- Build Your First MCP Server (Aug)
- The Quick Start to Dynamic AI Agents (Aug)
Advanced Builds, FinOps, and Custom Tools
- NotebookLM 2.0: From Research Tool to Creative Powerhouse (Sep)
- Building a Custom MCP Server for Gemini CLI: A Hands-on Guide (Sep)
- Exporting Your Google Cloud Billing Data to BigQuery (Sep)
- Build a Google Cloud FinOps Assistant Agent with ADK & MCP Toolbox (Sep)
- Build a GCP Cost Agent with ADK and MCP Toolbox for Databases (Sep)
October Surge: Workspace Integration & Deployment
- Create an AI-Agent in Google Workspace: Step-by-Step with ADK & Vertex AI Agent Engine (Oct)
- Deploy Your ADK Agent to Vertex AI Agent Engine: A Complete Tutorial (Oct)
- Turn Google Docs Into an AI Agent Hub: Integrate ADK Agents in Google Workspace (Oct)
SmartCheckbox Launch & Fun Experiments
- Getting Started with SmartCheckbox™: Install, Setup, and Your First Rule (Oct)
- Hands-On Tutorial — Build a Content Approval Workflow with SmartCheckbox (Oct)
- Integrating Zomato’s MCP Server into Gemini CLI: AI-Powered Food Ordering in Your Terminal (Oct) — A lighter, experimental piece that was genuinely fun to build.
The work I’ve been doing isn’t just content, it’s helping shape how teams and developers think about productivity, automation, and AI.
Product Collaborations & Developer Education
A meaningful part of my work this year involved collaborating with product teams in a way that closely resembled developer education and advocacy. My role was to explore their platforms hands-on and produce developer use cases, getting-started guides, tutorials, and documentation that simplified adoption.
This work helped bridge the gap between product capabilities and how developers and users actually experience them.
The Not-So-Perfect Side — And What I Learned
Not every moment in 2025 sparkled. At one point during the year, I lost a client I was writing content for. The work itself wasn’t the issue. The problem was communication.
I was traveling to another city for a community talk and didn’t communicate clearly about timelines. A small delay, combined with silence on my end, led to misaligned expectations. Eventually, the client decided to move on.
It wasn’t a dramatic failure — just a quiet reminder of something important:
technical skill and good intentions don’t replace clear, proactive communication.
Reflections: What 2025 Taught Me
Looking back with humility, this year was about consistent progress over perfection. A few grounded lessons that stand out:
Shipping real products (like the add-ons) creates momentum nothing else can — even if they start small.
Agentic AI shines when grounded in reliable data and clear actions; the MCP series reinforced how crucial live database connections are.
Teaching and presenting amplify impact — the live training and Mumbai Tech Week, Devfest, CCDs reminded me that sharing knowledge openly builds community and sharpens your own understanding.
Balance and resilience matter — graduating while maintaining a steady output wasn’t always easy, but celebrating small wins kept the momentum going.
I’m quietly proud of how far the work has come, but I also know there’s so much more to explore.
Looking Ahead to 2026
I’m excited to go deeper: more sophisticated multi-agent systems, production-grade deployments, expanded add-on features, and perhaps new courses or tools. The goal remains the same — build practical solutions that make Google Workspace and AI more accessible and powerful for everyone.
Thank you to every reader, installer, attendee, and community member who engaged this year — your feedback, questions, and encouragement make all of this worthwhile.
Top comments (0)