AI Agents: Learning by Experimenting
- Jinal Sanghavi
- Aug 21
- 3 min read
A few days ago, I wrote this post on LinkedIn that unexpectedly went viral, far beyond my expectations. This really made me pause and reflect.
We talk a lot about “catching up” in India, in tech, in our startups and companies. But what if catching up isn’t the point? What if the real shift is about how quickly we can leap forward?
AI agents aren’t just another wave; they’re a signal that the ground beneath us is already moving. The best part? It’s never too late to jump in, experiment, and learn alongside the rest of the world.
Curious to know more about AI agents? I think of them as intelligent automations, which responds to changing context/ acting assistants. It's LLMs on steroids! I found this intro on Agent.ai to be a particularly useful intro.
Particularly, it's important to understand that current technology isn’t yet at the point where you can simply define a goal and let the agent handle everything independently but it is increasingly getting there faster than we can imagine. AI agents exist on a spectrum of autonomy and below is a nice summary of how Agent.ai describes the levels.

Finally, very early days but I tried my hand at experimenting with Agents
Poppy Playwell on Agent.AI: A simple agent that sends daily emailers on a short DIY activity I can do to engage with my 4 year old daughter. Try it for yourself and let me know what you think
Automatic Gmail Email Labelling on n8n: Thanks to my cousin, found this to be a great template to get started with. This workflow automates email categorization in Gmail - It periodically checks for new emails, reads their content, and categorizes them based on existing Gmail labels. If no matching label is found, the workflow creates a new label and assigns it to the email.
Both Agent.AI and n8n have a bunch of templates to help get your started. And, there are a bunch of other websites like Make.com.
Continue to stay in the know. Below is the curated list I have found quite useful to keep pace with what's happening.
Follow on LinkedIn
Dharmesh Shah: Founder of HubSpot who started Agent.ai as a project, a place where you can find and connect with AI agents. Why? Because he's been writing software for 30+ years and never have has been more excited about the potential for AI
Saanya Ojha: Partner at Bain Capital Ventures who's frequent musings on what's up in the AI world are super insightful
Sangeet Paul Choudhary: Earlier, loved reading his takes on platform strategy. And, he recently released a new book Reshuffle: Who wins when AI restacks the knowledge economy - that has rave reviews and I plan to read over the weekend
Tune in to their podcasts
Dwarkesh Patel: This 23-year-old man from the Valley has built one of the most deeply-researched podcasts on artificial intelligence (and other other subjects), getting praise from the likes of Jeff Bezos, Mark Zuckerberg, Satya Nadella who have all appeared on his shows
Lex Fridman: a Research Scientist at MIT focused on human-robot interaction and AI research who's past guests have included Elon Musk, Jeff Bezos, Mark Zuckerberg. A lot of his recent popular episodes are around AI
Retailgentic: A niche podcast on exploring the intersection between retail, ecommerce and Agentic AI. Not very flashy speakers but quality content so far
Check out these newsletters
Lenny's Newsletter: Deeply researched weekly newsletters on everything AI and accelerating your career growth. Including a section on "how I AI"
Nate's Substack: Strategic takes on AI for news and views
Curious to learn your recommendations too! Who to follow, hear, read, etc.?
Comments