Neel Seth
I try to build and scale products that make money.
Product, growth, and business.
Learning in public, making mistakes, and building systems that work. I can't code without AI-but with AI, I can build almost anything.
What I Actually Do
Four areas where I spend my time solving problems
Find the Right Lever
Product, marketing, sales, partnerships-these are tools, not jobs. I figure out what actually moves revenue and focus there. At ShakesBierre, it was cutting platform spend and running my own ads. At Dukaan, it was enterprise onboarding. The lever changes. The goal doesn't.
Cut Through the BS
Platforms claim attribution they don't deserve. Agencies sell FOMO. Vanity metrics look good but mean nothing. I run my own data, question everything, and focus on what actually converts to revenue. Not growth hacking-just honest measurement.
Build What Makes Money
Features don't matter. Revenue does. I build the simplest version that solves a real problem people will pay for. ReplyDaddy: $11K in 90 days. Dukaan Enterprise: $1.2M ARR. Not because the product was perfect-because it solved a problem worth paying for.
Move Fast with AI
I can't code without AI-but with AI, I ship production systems solo. Python, Flask, React, RAG, agents. I'm not a 10x engineer. I'm a 1x product person who uses AI to do what used to require a team. Speed is the advantage.
Everything is a tool. Revenue is the north star. Pick the right metrics, pull the right levers, and do whatever it takes to move the number that matters.
Proof, Not Promises
Numbers that show what I've actually built and scaled
These aren't vanity metrics. They're proof that what I build actually works.
Work Experience
The jobs, startups, and experiments that shaped how I think about building products
Things I've Built
Products, tools, and experiments. Some succeeded, some didn't. All taught me something.
ReplyDaddy
AI-powered Reddit marketing tool. Built solo with AI assistance. $6k revenue in first 70 days. Currently growing at ~15% MoM. Proof that AI-assisted development works.
TeacherOp
Platform for teachers to run online classes. Built during COVID when teachers needed one tool instead of five. Still active with regular users. My first product that people actually paid for.
bot9.ai
AI chatbot builder for customer support. Led product and growth. Got press in TechCrunch and Insider. Taught me how to work with LLMs before ChatGPT made it mainstream.
Shipstation.ai
AI website builder. Reached 1,700+ users. Validated demand for no-code AI tools, but didn't nail retention. Learned a lot about AI UX and user expectations.
NoobBook
Educational NotebookLM clone built to teach AI/LLM integration patterns. Covers RAG pipelines, agentic loops, multi-modal AI (vision, audio, TTS), embeddings, and prompt engineering. Used in my GrowthX teaching sessions.
Dukaan Enterprise
Led product for Dukaan's B2B offering. Scaled from 0 to 10,000+ paying merchants in 7 months. Experimentation, onboarding optimization, and pricing strategy.
Rollout.site
Another AI website builder experiment. Faster than Shipstation but suffered from same retention issues. Sometimes you need multiple attempts to figure out what's wrong.
Delta Exchange Tools
Built internal tools for PM productivity and an AI trading bot. Client project. Worked, but not something I'd build for myself.
Trekkit
E-commerce for outdoor gear. Early project. Taught me the basics of building for transactions and inventory management.
Not every project was a home run. But every project taught me something I use today.
Learn how to build AI tools & Agents
I didn't plan to teach. I just shared how I was building with AI-and it turned into a free course.
From 2023, I was building products with AI tools-using ChatGPT for ideation, Claude for coding, and various APIs to ship faster. A friend at GrowthX saw what I was doing and asked if I'd teach their community how to use AI tools to build products.
I wasn't an expert. I was just someone who figured things out. But that turned out to be the point.
Most "AI courses" are taught by people who already know how to code. They assume you understand APIs, JSON, and terminal commands. I didn't come from that background-I learned all of this using AI as my coding partner. So I could teach it the way I actually learned it: messy, iterative, and focused on getting things to work instead of being "correct."
The first cohort had 50 people. Then 100. Then requests for more sessions. Now the free course materials on GitHub have been used by 150+ people learning to build AI tools without a CS degree.
People who've used the course materials
6+ hours of content, all free
No courses to buy, no upsells
What You'll Actually Learn
No fluff, no theory-only lectures. Just practical skills to build real AI tools.
API Fundamentals
Learn how to talk to AI models (Claude, GPT, etc.) via APIs. No prior coding experience needed-we start from absolute basics.
Build Chat Applications
Create conversational AI apps with persistent context, multi-turn conversations, and real-time responses.
Multi-Agent Systems
Build systems where multiple AI agents work together-one researches, one writes, one formats. Orchestration at scale.
Production-Ready Tools
Move beyond toy projects. Learn to handle errors, manage costs, and deploy tools people actually use.
Available Sessions
Full recordings and code samples-completely free
Introduction: The AI Learning Roadmap
A complete roadmap for learning to build with AI. Covers the key concepts, tools, and progression path you need to go from zero to building production-ready AI tools and agents.
Topics covered:
Session 1: API Fundamentals & Business Applications
Master the foundations of AI tool development with hands-on API implementation, moving from basic calls to production-ready solutions that solve real business problems.
Topics covered:
Session 2: Chat Applications & Advanced Agents
Build sophisticated chat applications with persistent context, multi-agent architectures, and production-ready features including real-time updates and specialized sub-agents.
Topics covered:
Session 3: Building NoobBook - A NotebookLM Clone
Build a complete NotebookLM clone from scratch. Covers RAG pipelines, multi-modal AI (vision, audio, TTS), embeddings with Pinecone, agentic studio generation, and full-stack AI application architecture.
Topics covered:
How It Works
All sessions are free and available on GitHub. No gatekeeping, no paywalls.
What Participants Say
Building AI Coworkers @ OneDigital | Founder SmartCue | Enterprise AI
“Should have better structure Start from the basics Went into nitty gritties before setting right context etc.”
Co-founder AnalyticsVerse | Enabling Constant Improvement for Dev Teams | Ex-AWS Ex-Arista
No written review
Senior Software Engineer I at QuantamBlack AI by McKinsey | Cloud & AI | ex-WatchGuard Technologies
No written review
Software Engineer
“First of all, the session was full packed with full of knowledge, the speaker started with very fundamentals that are not mostly covered when teaching such topic, but after setting up the fundamentals, we went and dived into the code and how the system actually works, the tools, the prompts and what not, but if you ask me what part of the session i enjoyed the most, then if i had to pick one, it would be the part where the speaker showed how the real world startups are using this LLM's to solve the problems at the scale, i am very excited to learn on such topics, hope we get more such sessions.”
Building Games x AI
“Neel clearly had a lot of knowledge to share, but overall the session felt a bit unorganized in terms of flow. it felt both easy and tough to follow, at different points of time. I wasn't sure what to take away from today's session. I am pretty sure if the material was organized and prepared with more time in hand, it would have been way better. I would love to attend a follow up session with the content and flow more organized and easy to follow.”
Senior Marketing Advisor Homefeild
“Idea: Build a small live tool during a meeting - something simple but useful, like a blog topic generator or keyword search helper. Goal: Show how quick and practical tool-building can be in real-time.”
Senior Analyst - Software Engineer Travelex
“The session was really helpful. Got to learn more about system prompts, tools and how we can have one tools talking to another. Looking forward to the next session where we can continue on exploring the usecases for automations using AI. Thanks again for sharing the valueable insights on building AI tools.”
Advanced Analytics and Automation Consultant Dropoff
“First of all, thank you for taking the time to share your learning. It was a very useful and insightful session. The content was extremely well organized, and I especially appreciated the hands on approach. Personally, I find it much easier to understand code than documentation, so having your examples cover the basics in such depth really helped clarify the finer details of working with LLMs. If I may suggest, it would be great if future sessions could start on time, considering how valuable both the hosts' and participants' time is. Punctuality helps everyone make the most of the learning opportunity. Also, given the richness of the material, it might work even better if the content were split into two 2-hour sessions, spaced about a week apart. This would give participants time to experiment with the initial steps before returning for a more hands-on continuation. I'm looking forward to trying out the code myself. It's truly a time saver when an expert, especially one self trained, takes the time to explain the logic behind their work so clearly. Thank you once again for an excellent and engaging session.”
Amit Goyal
No written review
Manoj Chaudhary
No written review
Amith Sasi
No written review
Product & Growth @ Kirana Club
“Have attended both session. Very helpful.”
Vaibhav Bansal
No written review
Leading Growth & GTM StreamAlive
“This was an amazing session, neel. Easy to understand and highly actionable. For the next session I'd love to learn more about langgraph and how to build complex agents and deploy it.”
AVP Brand Marketing and communications Upwisery
No written review
Independent Software Consultant
“Neel is great at explaining complex tech jargons in a simple manner that could be understood by anyone, no matter their skill level. Session was full of practical knowledges from his product building experiences and progressed a comfortable manner from easy to complex techniques without making overwhelming everyone. However more hands on coding could be added to make the session more interactive.”
Practice Lear Data Science & Analytics Mindstix Software Labs
“Neel is amazing. Content was absolutely on point. One of the most important takeaways: He made it super simple to try out a lot of stuff without over-engineering.”
Advanced Analytics and Automation Consultant Dropoff
“Thank you Neel for the fantastic sessions. We've covered the fundamentals of using LLMs in code quite well so far. It would be great to now see a live demo on how to use prompts (vibe coding) to build an agentic application and take it all the way to production.”
Product Manager Kratos Gaming
“Went through the 1st Session, what an eye opener. Kudos Man. Thanks for taking these. Wish it were slower, hand-holding with an absolute execution from Level-wise Topic to Prompts Creation to Execution to close every session for an absolute beginner like me but can't ask more. Thanks a lot again.”
Global Solution Architect Informatica
“Thank you for delivering such a wonderful and insightful session. You covered complex material exceptionally well within the time allotted. The content provided invaluable clarity on the tools, multithreading, and, most importantly, how to best leverage these concepts to solve problems across various verticals. This knowledge is immediately actionable. We genuinely appreciate the effort and expertise you shared and look forward with enthusiasm to your upcoming sessions.”
Lead Product Manager Sequioa
“Really enjoyed the last Agentic AI session. It was refreshing to see something so practical and grounded instead of the usual toy demos. I'm especially curious to learn more about the real-world optimization ideas you mentioned.”
Yash Yerma
No written review
Piyush Shah
No written review
Practice Lead Data Science & Analytics Mindstix Software Labs
No written review
Sandeeip Chincholkar
No written review
Chetan Parmar
No written review
Vedant Patil
No written review
Chetan Parmar
No written review
47
Total Reviews
14
Written Reviews
34
5-Star Reviews
4.6/5
Average Rating
Education
Github Activity
Some of my public repos, live projects, practice code and small experiments!
Contribution Activity (Last 6 Months)
Featured Projects
Shipstation AI
AI-powered personal portfolio website builder. Generate and manage your portfolio with natural language using Anthropic AI, React, and Node.js
NewsTracker
Unbiased AI-powered news tracker using Tavily and Anthropic. Tracks topics, analyzes sentiment, fact-checks information, and generates comprehensive summaries
PPT Daddy
AI-powered presentation generator using Anthropic Claude. Creates beautiful HTML slides and exports to PPTX format with custom branding support
BlogBuilder
An automated blog generation tool that creates SEO-optimized content
Teacherop(Neel) Free AI Course Powered by GrowthX
Free comprehensive course teaching non-developers how to build AI tools with Claude's API - from beginner to advanced with practical examples
Writings & Thoughts
Sharing insights on technology, entrepreneurship, and product development
Mental Modal Before Prompting The Right Way
A guide for prompting from personal experience. Preparation for the prompt is 10x more important than the prompt itself.
Building with AI: A Practical Roadmap for Builders, PMs, and Founders
Understanding inherited decisions, the cost of not knowing what you're building, and a practical roadmap to build with AI tools effectively.
Interesting Times Ahead
Personal analysis on the AI SaaS landscape, vibe coding tools, and why MCP/standards will win. A deep dive into Cursor, Lovable, Replit, and the vanity metrics problem.
The Real Cost of SEO: A Practical Guide to Ranking in 2025
Everything you need to know about SEO, backed by real data and actual costs. A comprehensive breakdown of strategies, budgets, and realistic timelines.
The End of UI: Why We Won't Need Traditional Interfaces Anymore
Understanding why UI was a workaround for technological limitations and how LLMs are making traditional interfaces obsolete.
Skills
Learning how to code, While Vibe Coding
Using llms
Product Management
Business & Growth
How I Got Here
The non-linear story of learning to build products that work
This is the messy, non-linear story of how I got here. No perfect trajectory-just consistent effort and a willingness to look stupid while learning.
Let's Connect
I'm always open to discussing new opportunities, collaborations, or just having a great conversation about technology and innovation.