I Was Wrong About AI Consulting (what I learned)
Key Takeaways
Shaw Talebi shares his experience of quitting a corporate data science job to become an AI consultant, highlighting the challenges of sales and non-technical aspects of consulting, and the importance of trust, discovery, and finding one sales channel. He also discusses his journey of generating revenue through YouTube, Medium blog, and paid consulting calls, and growing his YouTube channel.
Full Transcript
last year I quit my corporate data science job to pursue entrepreneurship full-time my plan was to sell data science Services as a way to fund the development of a product I could build a business around while this made a lot of sense on paper pursuing this path over the last N9 months has made me realize this plan was flawed in this video I'm going to share my experience and some key Lessons Learned in case it is helpful to anyone on a similar journey and if you're new here welcome I'm sha I make videos about data science and Entrepreneurship and if you enjoy this content please consider subscribing that's a great no cost way you can support me in all the videos that I make right out of grad school I went to work as a data scientist at Toyota this was in many ways my dream job and an incredible learning experience for me however after about 6 months in the role that initial excitement and learning curve began to flatten out and I slowly began to realize that the role was no longer aligned with my longer term goal of running my own business so after about a year in that role I decided to pass on a senior data scientist promotion and tank my income from over 10K a month down to basically zero since I had done some freelance work in grad school and had grown a small audience on YouTube my plan was to bring these things together and leverage my content to sell consulting services and to my surprise it worked over the next 8 months I took 36 Discovery calls of these 36 calls two of them turned into contracts and last month one of these contracts turned into an even bigger opport opportunity of over $25,000 where I was sitting in a project manager role and not doing any of the coding myself while it may sound like things were going great something was off this was similar to what I felt facing the promotion at Toyota it was a great opportunity on paper but something about it didn't feel aligned with my long-term goals so I made the tough decision to pass that opportunity off to another consultant looking back it's clear that my expectations of Consulting didn't match reality when I started this journey I saw Consulting as an easy way I could make cash while I explored other Ventures however after pursuing it as my main source of income as opposed to a side hustle like I did in grad school it became obvious that running a Consulting business wasn't as simple as I expected not just because of the technical challenges of building AI projects but also selling yourself nurturing leads working with subcontractors and the list goes on and on in fact most of the work were these non-technical aspects of the job with the biggest piece being the sales process as I've learned there are many unique challenges in selling AI Services three of which are as follows one for most businesses AI is a nice to have rather than a musthave so a lot of times it's not the client's number one priority two building AI projects requires a lot of experimentation and iteration which introduces a lot more uncertainty than the traditional software development process and reduces the perceived value of your offer and three since these are typically High ticket contracts they often require multiple touch points with the client before they close and I found this extra time commitment difficult to manage as a solo operator although I was learning a lot Consulting was taking up much more of my time and attention than I had anticipated so much so that my content output began to slow down and I virtually had no time to work on my own projects which was supposedly the main goal of all this this experience led me to take a step back and reminded me of some advice I had received from a successful product entrepreneur about a week after quitting my job I had asked him if Consulting was a good stepping stone to product development to which he immediately responded no the advice he gave was simple if you want to build a product then build a product looking back it's kind of funny that it took me 9 months to realize what he had told me 9 days into this journey but here's what I didn't fully appreciate building a product is hard building a consultancy is hard building a brand is hard entrepreneur ship is just hard the trick at least in my opinion is to pursue the hard thing that gets you fired up and that you find fulfilling and after trying it for 9 months I realized that selling AI projects to clients didn't get me as fired up as some of the other things I was working on that's why last quarter I removed the discovery call option from my website and passed that first major contract off to another consultant although building a consultancy wasn't for me I still believe it's a great business for those who enjoy it it also taught me a ton about sales marketing and working with customers which are universally applicable skills and Entrepreneurship if I had to boil it down here are my four key takeaways from this experience first is trust is more important than anything else for me what differentiated clients from prospects was the belief that I could solve their problem and that I was on their side through a lot of trial and error I eventually landed on the following approach be curious be transparent and be yourself more specifically be curious about the client's problem and where they're coming from be transparent about the limits of my skills and knowledge and to just be myself not trying to put up a front and pretend to be something that I'm not the second takeaway was not to skip the discovery when providing Technical Services like data science it's easy to dive head first into the coding the problem with this is that people end up spending a lot of time and energy solving the wrong problem that's why at the outset of every project it's critical to put on your project management hat so you can understand the business problem and fully scope a proposed solution the third takeaway is to find your one sales Channel although there are countless ways you can get clients upwork Fiverr cold Outreach LinkedIn content creation speaking at conferences referrals and the list goes on and on I and most of the people that I've interacted with in the space have just one main lead source for me my main source was my YouTube channel Channel and my funnel looks something like this someone would watch a YouTube video book a discovery call after the discovery call we would do a paid Discovery phase where the goal was to get a clear understanding of the client's problem and to scope out the project requirements and goals following the paid Discovery is building a proof of concept and then after the POC building an MVP the fourth and final takeaway is that it's not real until the money's in your bank account this is a lesson I had to learn over and over again and maybe I still haven't learned it there were many times I would have a great discovery call or multiple calls with prospects and it seemed like they were ready to move forward but then days and weeks would go by and I wouldn't hear from them and so while there's always excitement in sales I had to adopt this mindset to avoid going on these weekly emotional roller coaster rides at this point you might be thinking sha if you're not selling your data sign skills how are you going to make money while contract work has great short-term earning potential it is not my only Revenue source there are three other ways I've generated Revenue these past 8 months this includes revenue from my YouTube channel my medium blog and ad hoc paid Consulting calls which have generated a total of $766 38 although this isn't enough to pay the bills there's another thing here that's worth taking into consideration since quitting my job my YouTube channel has grown from 2,000 subscribers to 18,000 subscribers along with that my revenue from YouTube went from $100 in the first 3 months to 1,600 in these past 3 months which brings me to my new plan post one YouTube video a week while this might sound like an overly simplistic and also super risky plan here's my reasoning one YouTube is actually working for me two it allows me to focus on one thing three making one video a week gives me a clear quantifiable goal I can use to structure all of my efforts for instance here's a list of things that can go into making a YouTube video reading papers writing medium articles writing code examples talking to people conducting interviews building projects workshopping content ideas on other social media platforms and probably a lot more now here's a list of things that can result from making a YouTube video learning a new skill or topic getting more paid calls more speaking gigs more inbound leads more people joining the data entrepreneurs more content from my other channels and growing my audience nevertheless committing to one thing is scary especially something unpredictable like YouTube however the longer I spend on this journey the more I realize that commitment and focus are necessary ingredients for Success because this is the only way that every ounce of your effort can go in the same direction and to quote a fellow entrepreneur and friend Michael Lynn if you're doing less and less that means you're going in the right direction and indeed this feels like the right direction at least for now 9 months into this entrepreneurship journey I have three Reflections that are top of mind the first is I could have a very successful Consulting business and I could have a very successful YouTube channel but I can't have both I have to pick one and personally I just like making YouTube videos more the second is a subtle mindset shift which is instead of asking yourself will this thing work ask yourself how could I make this thing work it may seem like a subtle shift but this is the mindset that I'm adopting this quarter in making YouTube my main focus and the third and final mindset is to trust yourself trust that you'll figure it out trust that if you're backed into a corner your survival Instinct will kick in and you will solve the problem thanks for watching to the end I hope you got some value out of this if you have any specific questions about my journey feel free to drop them in the comment section below and as always thank you so much for your time and thanks for watching
Original Description
🤝 Work with me: https://aibuilder.academy/yt/INlCLmWlojY
🚀 Ship AI apps in weeks, not months: https://aibuilder.academy/courses/yt/INlCLmWlojY
I quit my full-time data science job to become an AI consultant. Here's how it went.
Introduction - 0:00
Background - 0:37
Expectations vs Reality - 1:55
4 Key Lessons - 4:24
How will you make money? - 6:50
Reflections - 8:59
Watch on YouTube ↗
(saves to browser)
Sign in to unlock AI tutor explanation · ⚡30
Playlist
Uploads from Shaw Talebi · Shaw Talebi · 0 of 60
← Previous
Next →
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
biometricDashboard2 DEMO
Shaw Talebi
biometricDahboard3 DEMO
Shaw Talebi
Time Series, Signals, & the Fourier Transform | Introduction
Shaw Talebi
The Fast Fourier Transform | How does it (actually) work?
Shaw Talebi
The Wavelet Transform | Introduction & Example Code
Shaw Talebi
Principal Component Analysis (PCA) | Introduction & Example (Python) Code
Shaw Talebi
Independent Component Analysis (ICA) | EEG Analysis Example Code
Shaw Talebi
Kmeans-based Blink Detecter DEMO
Shaw Talebi
Shit Happens, Stay Solution Oriented
Shaw Talebi
Why Conflict Is Good & How You Can Use It
Shaw Talebi
Causality: An Introduction | How (naive) statistics can fail us
Shaw Talebi
Causal Inference | Answering causal questions
Shaw Talebi
Causal Discovery | Inferring causality from observational data
Shaw Talebi
How to Be Antifragile | 7 Practical Tips
Shaw Talebi
Multi-kills: How to Do More With Less (no, not by multi-tasking)
Shaw Talebi
Topological Data Analysis (TDA) | An introduction
Shaw Talebi
The Mapper Algorithm | Overview & Python Example Code
Shaw Talebi
Persistent Homology | Introduction & Python Example Code
Shaw Talebi
What Is Data Science & How To Start? | A Beginner's Guide
Shaw Talebi
How to do MORE with LESS - multikills
Shaw Talebi
Causal Effects | An introduction
Shaw Talebi
Causal Effects via Propensity Scores | Introduction & Python Code
Shaw Talebi
Causal Effects via the Do-operator | Overview & Example
Shaw Talebi
Causal Effects via DAGs | How to Handle Unobserved Confounders
Shaw Talebi
Smoothing Crypto Time Series with Wavelets | Real-world Data Project
Shaw Talebi
Causal Effects via Regression w/ Python Code
Shaw Talebi
5 Reasons Why Every Data Scientist Should Consider Freelancing
Shaw Talebi
An Introduction to Decision Trees | Gini Impurity & Python Code
Shaw Talebi
10 Decision Trees are Better Than 1 | Random Forest & AdaBoost
Shaw Talebi
Dimensionality Reduction & Segmentation with Decision Trees | Python Code
Shaw Talebi
How to Make a Data Science Portfolio With GitHub Pages (2025)
Shaw Talebi
My $100,000+ Data Science Resume (what got me hired)
Shaw Talebi
How to Create a Custom Email Signature in Gmail (2025)
Shaw Talebi
I Spent $675.92 Talking to Top Data Scientists on Upwork—Here’s what I learned
Shaw Talebi
Lessons from Spending $675.92 to Talk to Top Data Scientists on Upwork #freelance #datascience
Shaw Talebi
A Practical Introduction to Large Language Models (LLMs)
Shaw Talebi
The OpenAI (Python) API | Introduction & Example Code
Shaw Talebi
The Hugging Face Transformers Library | Example Code + Chatbot UI with Gradio
Shaw Talebi
Why I Quit My $150,000 Data Science Job
Shaw Talebi
Prompt Engineering: How to Trick AI into Solving Your Problems
Shaw Talebi
The REALITY of entrepreneurship. #entrepreneurship #startup #smallbusiness
Shaw Talebi
Fine-tuning Large Language Models (LLMs) | w/ Example Code
Shaw Talebi
How to Build an LLM from Scratch | An Overview
Shaw Talebi
I Have 90 Days to Make $10k/mo—Here's my plan
Shaw Talebi
I Spent $716.46 Talking to Data Scientists on Upwork—Here’s what I learned.
Shaw Talebi
Pareto, Power Laws, and Fat Tails
Shaw Talebi
Do NOT become an entrepreneur #entrepreneurship
Shaw Talebi
Detecting Power Laws in Real-world Data | w/ Python Code
Shaw Talebi
How I’d learn data analytics (if I had to start over in 2024) #dataanalytics
Shaw Talebi
4 Ways to Measure Fat Tails with Python (+ Example Code)
Shaw Talebi
Fine-tuning EXPLAINED in 40 sec #generativeai
Shaw Talebi
How Much YouTube Paid Me in My First 6 Months of Monetization (as a Data Science Creator)
Shaw Talebi
5 Questions Every Data Scientist Should Hardcode into Their Brain
Shaw Talebi
AI for Business: A (non-technical) introduction
Shaw Talebi
LLMs EXPLAINED in 60 seconds #ai
Shaw Talebi
3 Ways to Make a Custom AI Assistant | RAG, Tools, & Fine-tuning
Shaw Talebi
What is #ai? — Simply Explained
Shaw Talebi
QLoRA—How to Fine-tune an LLM on a Single GPU (w/ Python Code)
Shaw Talebi
How to Improve LLMs with RAG (Overview + Python Code)
Shaw Talebi
Text Embeddings, Classification, and Semantic Search (w/ Python Code)
Shaw Talebi
More on: PM Basics
View skill →Related Reads
📰
📰
📰
📰
The Best AI Course Generator in 2026? I Tested 7 to Find Out.
Dev.to AI
Automate Spotify and YouTube Playlists - Chapter 2: Setting Up Spotify
Dev.to · Tawanda Nyahuye
Use a model route manifest before Dify, Cursor, and Node.js share Vector Engine
Dev.to AI
What 90 Days of Comments on AI Side Panels Taught Me About Distribution
Dev.to · AI Buddy
🎓
Tutor Explanation
DeepCamp AI