A live cohort for product managers who are done reading about AI. You design it, build it, and launch it to a real URL — no engineering background needed.

This isn't a notify-me box. The waitlist gets you in first — and locks the lowest price of the launch.
When the cart opens 1 October, waitlist members get the link before it's public. Cohorts this small sell out from the waitlist alone.
The lowest price of the launch is reserved for the waitlist. You see the number before anyone else — it only goes up from there.
Full curriculum, frameworks, tools, format, and time commitment — all on this page and in your inbox.
Waitlist members get the raw view of how this cohort is built — plus a free framework or two before the cart opens.
By the final session you don't have notes and a PDF. You have a working product and the exact IP to talk about it in any room.
The same spine that took 80+ PMs from I-use-AI to I-shipped-AI — condensed to six focused weeks. Each week ends with a concrete deliverable that becomes part of your product and portfolio.
Think like an AI PM before writing a line of code — the 4 Pillars, the 7 Questions operating model, and designing for when the AI is wrong (trust, control, recovery).
Get fluent in the technical concepts every AI PM needs — tokens, RAG vs fine-tuning, embeddings, agents, and context engineering — without becoming an engineer.
The most valuable thing an AI PM does: confidently say yes — or no. The 5-Gate framework, the 6 AI Patterns, red flags, and how to teardown any live AI product in 20 minutes.
Define exactly what you're building and how you'll know it works — the 9-section AI-First PRD, quality examples, and the Eval Framework with LLM-as-judge.
Go from PRD to a working AI product, then make it reliable — vibe coding, observability, the Testing Loop, and the four launch audits (three fully automated).
Ship it, show it, own it — the launch-post and landing-page formula, demo craft, and presenting your live product to the cohort.
Developed and used in production before the course existed, then tested with 80+ PMs across two cohorts.
From what-problem-should-we-solve to how-do-we-keep-making-it-better — the decision model behind every product students shipped.
Probabilistic? Data ready? Pattern match? Error tolerance? ROI? Five gates that stop you building AI for problems that don't need it.
A 9-section spec built for probabilistic products — quality examples, eval dimensions, error handling, model-degradation contingency.
Dimensions → test cases → 1–4 scoring → LLM-as-judge → a Golden Set that defines great. The skill most AI courses skip.
Every product below was built and deployed by a PM in a past cohort — no engineering background, no code written by hand. Press play.
Unedited words from past cohorts — Demo Day and post-course feedback. Not cherry-picked one-liners.
“Sid's course design and his ability to meet you at your level — no matter how technical you are or aren't — was incredible. I got hands-on experience with product planning, vibe coding, testing my product, and deploying it to a live URL.”
“My French is better than my SQL and I don't even know French. For a person like me with not a lot of notion of tech to be building something from scratch — it's all thanks to this course.”
“I was immediately able to create a prototype that's getting passed to designers at work. It had immediate value — and I got put on higher-visibility projects directly after that because of it.”
“I didn't know that AI products I would be able to build in less than two weeks. The best part is the pre-reads are genuinely helpful — and Sid answered me even in his late-night hours.”
“I personally learned a lot and applied some of these ideas at work. The theoretical depth was super helpful — it's what helps you get it right and avoid fundamental mistakes.”
“I've never been a developer in my life — not a single line of code. I never believed I would build something. Thank you for instilling that belief and trust.”
“No professional achievement came as close as the joy of building the app on your own — that too for a problem so close to your heart. Taking that first step was what really mattered.”
“Before this course I felt scattered — bits and pieces of knowledge, no coherent framework. It brought structure to everything and gave me a durable way to think about AI product management.”
“You bring the right blend of theory and practical knowledge — that's been the best takeaway. Every single person in this cohort talks about how well you supported them. That speaks for itself.”
“I really appreciate the structure you bring to the teaching — it helps people like me learn on the side. I've been able to flex some of the knowledge at the office and get real benefits from it.”
One live session a week, recorded within 24 hours. Plan for about 4–6 hours a week: the live session, short pre-reads, and building.
One a week, led by Sid — teaching, product teardowns, hands-on activities, and open Q&A. Every session recorded.
Short, practical reads (10–20 min) before each session that give you the vocabulary to get the most out of the live time.
Every session ends with a concrete artifact — a teardown, a PRD section, a test set, a working thin slice. These are your product and portfolio.
Live for the whole program and beyond. Sid is active between sessions. Build help, PRD feedback, peer review.
At least one practitioner with direct production experience — a past cohort had a dedicated evals session from someone who ran evals at scale.
Every recording, framework, template, and pre-read — plus the cohort community across every cohort, forever.
The lowest price of the entire launch is reserved for waitlist members — revealed the moment the cart opens on 1 October, before it goes public.
Cohorts this small sell out from the waitlist. Join now and when the cart opens 1 October you get the link first — at the early-bird price that only goes up from there.
Join the waitlistNo. The course uses vibe-coding tools — you describe what you want in plain English and the tool builds it. You manage direction, quality, and testing. Being comfortable reading code helps, but you'll never have to write it.
The cart opens 1 October to 35 seats — waitlist members get the link first. The cohort starts 10 October and runs 6 weeks.
Three things: first access to the cart before it's public, the lowest (early-bird) price of the launch, and every detail — full curriculum, behind-the-scenes on how the cohort is being built, and a free framework or two before the cart opens. Joining is free with no obligation to enroll.
The waitlist locks the early-bird price, revealed when the cart opens on 1 October. Joining the waitlist is free — you'll see the number before anyone else and decide then.
Roughly 4–6 hours a week: ~1.5–2 hours live, 30–60 minutes of pre-reads, and 2–3 hours of building. The build weeks are the most intensive. Every session is recorded, so you can catch up.
Past cohorts ran sessions that worked across IST, GMT, and EST/PST. Final scheduling is confirmed as enrollment closes — timezone preferences are collected at signup, and every session is recorded either way.
No, but it helps. There's a structured ideation exercise to find a buildable idea — a problem you have, a narrow user, a clear input and output. If you already have one, you'll pressure-test it against the 5-Gate framework.
A live, deployed AI product at a real URL; a completed AI-First PRD and Eval Framework; a 2–3 minute demo video for job applications; and the frameworks (7 Questions, 6 AI Patterns, and more) you can apply at work immediately.
Fastest way is to message Sid directly on WhatsApp — he reads and replies himself.
Join the waitlist — first access to 35 seats, the lowest price of the launch, and every detail as the cohort comes together.
Join the waitlistStill deciding? Ask Sid anything on WhatsApp →