The AI Entrepreneurship thread. Code AI? Interested in AI? Data Science? Machine Learning?

[ChrisV]

Artificial Intelligence and Machine Learning are likely going to be the biggest boons of the 21st century.

Possibly one of the biggest technological boon the world has ever seen.

The Entrepreneurship opportunities are going to be endless.

I wanted to make a space for those pursuing AI or interested in AI to share resources and any type of AI related

So are you pursuing AI? What's your story? What interests you about AI? Have any resources to share?

Or feel free to just talk about anything AI or ML related.

@ZF Lee , @Strategery , @David Moyses

Feel free to tag anyone who you know that's involved in Data Science, Machine Learning or AI

So what interests you about the world of AI?

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[PedroG]

Subscribed. I've been a software engineer for a long time but I'm fairly new to ML. I'm currently doing the Deep Learning specialization on coursera which is excellent. I have one more course to go. There's 5 total.

 

[ChrisV]

He took real-life problems that spiraled out of control (eg newspaper reports that miscommunicated data and led to government F*ck-ups and societal backlash) and showed us how the concepts come into play.

So I’m in the field more for the problem-solving, and the background story, that gives data its meaning, rather than the numbers or lingo themselves.

Yes! Statistics is boring in and of itself. It's what you can do with Statistics that's exciting.

The analogy I like to use is... Math is boring, right? Physics is boring, right?

But what if you could use your knowledge of Math and Physics to create the fastest Ferrari the world has ever seen. Now it's not so boring now is it? So the fields are always boring in a vacuum, but they become exciting when you see what they can help you accomplish.

Another great example comes from my favorite movie: Moneyball.

Stats is boring right? But in the early 2000s Harvard Statistician Paul DePodesta teamed up with Billy Beane and the Oakland A's to use statistics to create one of the most impressive baseball rosters of all time. They almost beat the Yankees in the World Series, but did it for one eighth of the price. Just by using Stats.

I made a quick 3 video Moneyball Playlist that I think illustrates everything nicely, even for people who have seen the movies

Human decision making is flawed because of a number of biases. Stats eliminates those.

In the Moneyball situation, certain players were overlooked becasue they were funny looking, had an ugly girlfriend, had a weird pitching stance. So most teams didn't realize how amazing they were as players because human decision making is biased.

 

[ChrisV]

My story: I run a Data Science company. I work with a number of researchers and various corporations to help them solve problems accurately through data. I'm hoping to expand my skills to encompass more Machine Learning techniques into my research. I got into Data Science becasue I just love Science. Not so much chemistry, or physics or anything else when you think of "science," but just the scientific method. The philosophy of science. The idea that rather that rely on intuition or faith, we actually test our ideas to make sure they're true.



I've found that with good data you can accurately answer pretty much any question you can think of.

Data Science gave me the opportunity to basically run quasi-experiments finding the answer to almost any question I needed to. Traditionally with science, it was a lot of work testing a hypothesis. You'd have to set up experiments, conduct long questionnaires, etc. Not so with Data Science.

From a Medium post I wrote a while back:

The world is changing. All sorts of new technology is on the horizon. Ever wonder how Amazon knows exactly what books you might like? How Pinterest figures out which ‘pins’ might interest you. How Apply Music or Spotify finds new songs that you absolutely fall in love with. How do these machines all this out? How are our devices becoming so intelligent? The answer is Data Science.

We now have the ability to track every click, swipe, ‘like,’ and repost that we want.

Within this data is enormous insights about consumer behavior and even general psychology. Companies have enormous amounts of Data and can do all sorts of incredible things with it. Data Scientist: The Sexiest Job of the 21st Century. Data Science can call presidential races, can tell you which movie is best for you based on your tastes and moods. Netflix upgraded itself from a simple DVD Mailing service and made itself a household name by applying algorithms that figure out which movies you’re most likely to enjoy.

Data Science can show you new music that you would like based on your tastes and even mood. Mercedes-Benz for instance is experimenting with heart-rate sensors to determine your mood so it can decide what music to play for you. It can tell if you’re in a relaxed mood and just want some relaxing music, or if you’re in an upbeat mood and want something more energetic.

And here's the full Medium post for anyone interested.

What is Data Science, and why should you care?

The world is changing. All sorts of new technology is on the horizon. Ever wonder how Amazon knows exactly what books you might like? How Pinterest figures out which ‘pins’ might interest you. How…

medium.com

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[ChrisV]

Hey Chris, thanks for sharing all this valuable info. What's your take on data sci vs finance? I currently work in real estate banking but am torn by the strong need for data sci/analytics in the market. Looking to maximize my job income as I work on a fastlane product in a separate field. What's your opinion on pursuing data sci vs finance? I intended to make real estate my career, (while working on side biz) but the need for data sci seems so strong. I dont usually believe in followings one's passion(over mkt demand), so still on the fence. My goal so far is to pick up using R/Python since both fields can sometimes mesh. Any sort of feedback would be appreciated!

1. What are you good at
2. What do you enjoy doing
3. What does the world need
4. What can you get paid for

We know that DS is something the world needs and is something you can get paid for... but what do you think you'll be better at? And which would you enjoy more?

I wouldn't just go do data just because it's a trend right now.. i'd think about what you're also suited for. And there's no reason you can't use Data Analytics in finance. Data can be applied to everything. I see no reason why you can't do both.

 

[adaaaam]

Hi all, just wanted to chip in as someone who works in this space. (Data Scientist)

For those considering getting into the field, I'd recommend considering this article (especially point #8). Don't take it too literally - but this echos a lot of the commentary I've personally made about the direction of the field.

Don’t Become a Data Scientist

The advice I give when someone asks me how to get into data science. Become a software engineer instead.

towardsdatascience.com

This is pretty much spot on, based on my experience. After a few years bouncing around data science roles plus freelancing, it looks to me like many companies don't really have use for AI/ML as-promised. And often they're lacking the infrastructure to support model deployment even if they have a use case. Of course they don't know that yet, so it's a great job market now. This might mean a near-term bubble.

But as companies advance their tech (largely: get on cloud!), there may well be a point where machine learning becomes relevant for these companies. But I expect the data science landscape to have changed largely by then. The statistics side of this is already all but irrelevant for the practitioner. That matters for the people at Google/Amazon/Microsoft developing the algorithms; but for everyone else, you can call them as an API from your favorite cloud provider. This means (IMO) that the "data science" many have learned (fitting models with sklearn) will be irrelevant, and a developer skillset more and more important.

Anyways, exciting times. I'm working on a "prediction-as-a-service" type of product now that will hopefully lower the barrier to entry to machine learning for smaller companies who are used to working in Excel.

Excited to hear more about what you guys are doing in this space and potentially partnering up!

 

[ZF Lee]

Hey Chris, thanks for sharing all this valuable info. What's your take on data sci vs finance? I currently work in real estate banking but am torn by the strong need for data sci/analytics in the market. Looking to maximize my job income as I work on a fastlane product in a separate field. What's your opinion on pursuing data sci vs finance? I intended to make real estate my career, (while working on side biz) but the need for data sci seems so strong. I dont usually believe in followings one's passion(over mkt demand), so still on the fence. My goal so far is to pick up using R/Python since both fields can sometimes mesh. Any sort of feedback would be appreciated!

(I forgot whether I included this nitbit into the lecture notes I sent you! @ChrisV)

Rob Hyndman popped by at my university the other day, and he described being a data scientists like this:
'Being a data scientist (or doing data science) is being more skilled in programming than a statistician, and being more skilled in statistics than a programmer.'

So, I feel the heart of data science is statistics, while programming for data science itself is the tool for the means to an end.

And if you look at the fundamentals of finance, such as standard deviations for risk and volatility, statistics is heavily linked to finance.

Me thinks some knowledge in just using Excel will just be fine.
I take finance units in university presently, and they don't need me to do R programming.

And just remember that any data analytics method can only be as good as its dataset.

If the dataset is too F*cked to be cleaned up...well, rubbish in, rubbish out.
This goes back to how the folks collected the data in the first place.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[akTwelve]

Great idea starting this thread! I don’t have any passion for sports, but I did want to chime in as an AI developer.

Rather than the data analytics side, I’m focused on Computer Vision and Reinforcement Learning using the Unity game engine. So far, I’ve created a cigarette butt detector and a weed detector (for my lawn), plus a number of projects where video game characters learn to navigate their environment purely by being rewarded for good behavior. I created a Udemy course on the technique I used to create the cig and weed datasets and I have a bunch of YouTube videos about the game stuff. If you’re curious, I post about it on my website, www.immersivelimit.com.

I’m particularly interested in the use of simulators combined with real world data to train robots and computer vision systems. I think there’s a lot of business potential there, especially since it’s so difficult. Definitely fits the Entry commandment of CENTS.

 

[ZF Lee]

Artificial Intelligence and Machine Learning are likely going to be the biggest boons of the 21st century.

Possibly one of the biggest technological boon the world has ever seen.

The Entrepreneurship opportunities are going to be endless.

I wanted to make a space for those pursuing AI or interested in AI to share resources and any type of AI related

So are you pursuing AI? What's your story? What interests you about AI? Have any resources to share?

Or feel free to just talk about anything AI or ML related.

@ZF Lee , @Strategery , @David Moyses

Feel free to tag anyone who you know that's involved in Data Science, Machine Learning or AI

So what interests you about the world of AI?

Haha I’m just taking some data analytics courses in university, and I would have never got into the field in the first place, if I didn’t have a great lecturer for my first year.

In high school, for elementary statistics, they just threw all the Greeks at you and you have to memorize everything without truly comprehending their roles.

But my lecturer was different.

He took real-life problems that spiraled out of control (eg newspaper reports that miscommunicated data and led to government F*ck-ups and societal backlash) and showed us how the concepts come into play.

So I’m in the field more for the problem-solving, and the background story, that gives data its meaning, rather than the numbers or lingo themselves.

Exploration is the key word.

For my current freelance copywriting jobs, I’m beginning to wonder if there were more effective methods of research, rather than just qualitative surveying and forum scrolling.

Or, if you will, I’m leaning towards a left-brained logical way of research, as sometimes leaving research to ‘wing it’ can result in some very deceptive interpretations.

 

[BlindSide]

Crazy thought I had: it’s estimated that every 50 minutes, someone is killed in alcohol-impaired accident. Could AI and technology be implemented into cars to sense an alcohol impaired driver?

Maybe it could prevent the car from being turned on? Or say it’s on the road and it senses driving habits that indicate an impaired driver, some type of ability for the car to pull over/get off at the nearest exit?

Food for thought for my crazy brain.. and the Elon Musks of the world. Just imagine: This is a $44B problem every year.

 

[ChrisV]

Great interactive Demo of how a Convolutional Neural Network works

CNN Explainer

An interactive visualization system designed to help non-experts learn about Convolutional Neural Networks (CNNs).

poloclub.github.io

Two more interactive Demos on Decision Trees:

A visual introduction to machine learning

What is machine learning? See how it works with our animated data visualization.

www.r2d3.us

A visual introduction to machine learning, Part II

Learn about bias and variance in our second animated data visualization.

www.r2d3.us

 

[ChrisV]

I just remembered I wrote this article a while ago:

The absolute fastest way to learn R for Data Science.

This post will be updated as times goes on. But for now, the most important things are:

medium.com

Data Science and Artificial Intelligence are distinct but similar fields, so the information in that article can also help you with AI / Machine Learning.





It's based on the R programming language, but many of the suggestions can be applied to python as well.

The best way to learn something is to learn the top ~20% of most important things, which will generally get you 80% of the results. In the article I suggest you just do it. Don't waste time learning a shitload of theory at first. You can learn that later. It's best to focus on the actual things that professionals use.



Jeremy Howard likens learning AI theory to learning baseball. Imagine if you wanted to learn to play baseball. But instead of just going out and playing baseball they sat you in a classroom and taught you the physics of the game. They taught you equations and theory and made you memorize it. This is obviously a really silly way to learn and by the time you got to even play a game you would have probably lost interest. He talks about it here.

How this relates to my article is: the best way to learn Data Science or Machine Learning is to do Data Science or Machine Learning. Kaggle.com has all sorts of datasets you can download and experiment on. That way by actually doing the work you learn the most important things used in production (the 20%.)

Some notable tips from my article:

Write the code. Don't copy and paste code you find in tutorials or StackOverflow. Make sure you actually physically type the code. You need the muscle memory. and you need feedback on what you're doing right or wrong. When programming it's so easy to put a period or closing bracket in the wrong place and unless you're actually typing the code you'll never know what you're doing wrong. Act like Copy and Pasting is a felony.

When you learn something: put it into practice immediately. Our cognitive memory is terrible, but out experiential memory is great.

Things need to be ingrained in your mind. You create different neural pathways when you do something. You don’t even necessarily have to intellectually know something to have the neural pathways. Have you ever needed to give someone a phone number, but without the number pad in front of you, you couldn’t remember it? When there’s a phone in front of you. you can dial the phone number just fine, but when someone asks you you’re like ‘uhhhh, let me go look at a number pad.’ Or the same with directions sometimes. You cant give someone directions but you ‘know the way if I just drove there.’ This is why doing is so important.

You have a muscle memory where you know exactly what you’re doing…. but you can’t intellectualize it. And the opposite can happen; you can intellectualize something, but have no idea how to actually do it. Don’t make that mistake!

You need to run the command immediately after entering to see if there are any errors. You absolutely need feedback. If you type it into the script, and don’t enter the command, you can easily engrain an incorrect syntax into your brain over and over.

Here are some evidence-based strategies for the best ways to learn things:



Btw, I have a bunch of life-improvement related articles on Medium as well.

Chris Vaccaro – Medium

Read writing from Chris Vaccaro on Medium. Evidence-based insights. Every day, Chris Vaccaro and thousands of other voices read, write, and share important stories on Medium.

chrisvacc.medium.com

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[The-J]

Don’t want to be a downer but there are a lot of people that work with AI and know it in and out, know what it can and can’t do and even they can’t create very profitable businesses.

That's the opportunity. Engineers aren't good business people by default.

For the layman, it's most likely a better idea to look for use cases and then seek those who have built tools that can do what you want to do. Entrepreneurs (unless they have the background to do so) don't need to be building the tools themselves; they need to lead the team that brings the tool to market. You think Elon Musk has mastered the mathematics behind computer vision? Absolutely not.

Entrepreneurs can lead a team of engineers to build the thing that they can't build themselves.

Also, I've spent the last few weeks reaching out to AI entrepreneurs and marketers and they're a LOT rarer than engineers working on AI. AI companies typically are staffed with dozens of engineers and many of them aren't profitable. Why? Because they don't have a profitable use case for it. But they have awesome technology! Lol

Also VCs are throwing piles of cash at companies with cool tech and a team of engineers, so money really isn't the barrier

 

[srodrigo]

So what interests you about the world of AI?

I love the endless applications it has. Literately, any thing can benefit from Machine Learning.

I was into ML for a while (before I got distracted with games programming) and loved it. But it felt like climbing the Everest. There's so much to learn that's one of the fields that really require specific education to get somewhere.

I played around with ML for finance and betting, it was fun. I have some real projects in mind, but they aren't my current priority. I specially like ML for real estate (although getting good data seems tricky).

 

[ChrisV]

Hey guys I have an idea if anyone would be interested.

Would any of you guys like to take part in the Fastlane Artifical Intelligence & Machine Learning Fantasy Football Competition.

Here's how it works...we basically use AI / ML Algorithms to create a Fantasy Football or Fantasy Baseball team, Moneyball style. I think since Baseball season is almost over, our best bet is Fantasy Football.

Even if you're not experienced, I think this would be an awesome project to apply analytics to a real life project. Even if you're not good at it yet... f--- it... no better way to learn than by doing.

Would anyone be down?

@PedroG, @ZF Lee, @Empires, @S.Y., @Strategery, @srodrigo, @spreng, @Fastlane Liam , and anyone else

 

[ChrisV]

Just one of the many terrifying things AI will be used for. I see a lot of potential for good too, but I think we will see a radical shift in our lives as the tech matures and spreads. Sentient AI is hardly a requirement either.

It's good in that it will be helpful in catching dangerous criminals, but it's bad in that we have to now trust the government to use it right.

Extreme example but: what if technology gets so good that they in the future they can scan somebody's face, get an estimate for their DNA, see if that DNA predisposes them to marijuana use, and then search their bookbag because their face said they might have genes more likely to be marijuana smokers.

Maybe an extreme example, but definitely plausible. I think with all this technology we're going to have to really go back and reassess if our laws are fair.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[ChrisV]

Just wanted to post some of my favorite resources for anyone who wants to learn,

This guy (quite ironically) used Data Science to rank every Data Science course on the net

The best Data Science courses on the internet, ranked by your reviews

by David Venturi The best Data Science courses on the internet, ranked by your reviews Photo by Quinlan Pfiffer [https://flic.kr/p/pra3h4]A year and a half ago, I dropped out of one of the best computer science programs in Canada. I started creating my own data science master’s program...

www.freecodecamp.org

But these are hands down my favorite courses on the net:

Analytics Vidhya

courses.analyticsvidhya.com

AnalyticsVidhya is hands down my favorite Data Science blog on the net.

Also for anyone interested, Harvard offers a real good soup-to-nuts Data Science program on their site. You can do it completely online

HarvardX Data Science Professional Certificate

Learn key data science essentials, including R and machine learning, through real-world case studies to jumpstart your career as a data scientist.

www.edx.org

 

[ChrisV]

I'm also compiling a list of my favorite resources in a separate thread. All the resources so far are free except Harvard's Data Science certification.



You don't even really need that though. I just got it because I thought it would be nice to add to a resume, which I havent even needed. In this field as long as you're capable and can do quality work, people will hire you. The formal certifications, etc. don't matter so much. In my opinion, you're better off showing off some cool projects of Github or whatever.

But anyway, these are some of my favorite resources:

My favorite Machine Learning / Data Science / AI resources

I'm gonna just use this as a resource dump for anyone interested. Feel free to add whatever you'd like. So I just realized that PBS's Crash Course (one of my favorite YouTube Channels) put up the beginnings of a full AI course on YouTube. View...

www.thefastlaneforum.com

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[ChrisV]

So guys, here's a fun fact. Have you ever wondered why those CAPTCHA puzzles always contain driving-related images? Well it turns out you're actually teaching Self-Driving cars how to drive.

Is reCaptcha Training Robocars?

Filling out a reCaptcha form makes you identify things you'd see on the road. Are you teaching Waymo's driverless cars? We investigate.

www.ceros.com

You are building a self driving AI without even knowing about it | Hacker Noon

hackernoon.com

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[Deleted74925]

So guys where do you see industry going in the upcoming years? Is it worth to dig deep into AI/Data Science if you will want to create something (considering you aren't a huge company like Google, Amazon etc.) and you're building from a ground up. Or is it better to go more into creating hardware/IOT stuff that will be needed for computation and also to gather data? If you would decide now to study one of them, which one you would go to?

The "frontiers" of AI are light years ahead of what you and I can access from online courses. Source: I manage an online course and it's taken me months just to create tutorials. I've heard of a few successful AI startups but they mostly get VC funding and have founders who went to Ivy league schools.

A more "fastlane" approach is to feed the crowd. What are the needs that all of these AI companies and people will need? All of these are unattractive, which means that there's less competition.

Skills

Sales and marketing of machine learning models

• How can you prove that your model is worth $100/customer? Data scientists tend to come from technical backgrounds. I worked as a predictive modeler who would create the models and PowerPoint decks, but then we'd turn them over to our sales team who would sell them to clients.
• My current course has about 30% of the curriculum just on communication. Basic things like writing, combining graphs/statistics into understandable language. The person who can explain their algorithm/AI wizardry in simpler language is worth more money than the person who struggles with communication.

Risks to fraud, data breaches, technical errors

• Data cleaning, integrity, and reducing error. How can you know that your data is going to be consistent? "Garbage In; Garbage out." Data scientists need to know this, but it's not new! It's been around in Database language since SQL was invented in the 1980s... but people don't want to learn it because it's "boring".
• Writing documentation.... Ugh. Is there anything that a data scientist hates more than doing this? Probably not. This task tends to get outsources to junior team members because it sucks. But the probme is that only the "expert" who created the model is knowleagable enough to write good documentation.

Hardware

• How to use cloud services. There are basically three big competitors - Microsoft Azure, Google Cloud, and Amazon Web Services.
  • Companies concerned about data privacy still are hesitent to use the cloud, unless they are massive and have boat loads of money to spend on R&D.
  • Getting models into "Production". It's easy for someone to take their Python or R and make some AI tool, but it takes a lot longer to get it ready to be used in real-time. Most of the time this needs to be translated from R/Python into some other framework like JavaScript/Scale (?), Julia, etc. This is one of those tasks that data scientists hate.
  • Budgeting... The cloud offers "unlimited scalability" but with that comes unlimited costs. At my former company, we ran into major problems with cost overruns. We were using a machine learning model which was designed to run on about 100 GB of data, but then we had a client that had 100TB of data. The cloud computing costs went from about 20% of the costs (for normal clients) to 60% of the costs. This could have been avoided if we had knowledge of how to manage costs.

Hiring/Staffing/Recruiting

• How do we hire the person with the right credentials?
  • Hire only Phd's
  • Hire Master's degree people and hope that their degree is actually useful
  • Put them through a rigourous hiring process involving technical projects, background checks with previous managers, reviews of their github/kaggle projects, and numerous on-site interviews and "white board" questions
• How do we know if a "data scientist" from India who is selling themselves on Upwork for $500/hr is legit or not?

Education

• Analytics managers have "continuing education" requirements like doctor's do. As algorithms get more complicated, it becomes more difficult to keep staff up to date. There's a growing need for corporate training. Remember all of that "Data Privacy" training that you were required to take? Imagine "AI Privacy" training in 3 years.
• Learning platforms - My site is build using Moodle, which is good because it has a ton of open source documentation.. but there are newer platforms which have more features such as zoom integration, templated content (think Wordpress) for quickly making new courses, and so forth.

 

[Martin.G]

Nice resource you share @ChrisV.

I am reading Reinforcement learning: an introduction (Richard S. Sutton, Andrew G. Barto), a great book that cover all what you want to know about the subject, from the beginning of this discipline to today.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[WillHurtDontCare]

The other option is a Master called "Data Science and Entrepreneurship",

You know what would be better than doing this?

Reading Cas$hvertising, getting clients, then learning how to use data science tools to solve specific problems that make you money.

You don't need a degree to use data science tools. Take OCR (optical character recognition) for example. You could spend time learning all of the ideas that went into building it, OR, you could learn how to use the Tesseract API with Python to data mine some text out of image documents and check if you got the results that you wanted. People will pay you to do this.

Graduate school is the new seminary - you don't go there to make serious money; you go there to be indoctrinated into the priestly caste and so that people acknowledge you for being wise.

You know what is way better than spending money to learn? Making money to learn! And you will learn data science 100x more effectively if you're solving REAL business problems.

Join some professional networking group like BNI and apply as an analyst of sorts. If you have an undergraduate in mechanical engineer, people already assume that you're smart. From there you can start researching the industries and professions of the people in your group, then once you understand their problems, you can start to offer potential solutions.

And just to summarize this post - you know that you can start working on a problem before you have the full solution, right/

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[Raja]

I'm looking for something a non-technical person can understand.

you will need to understand technical stuff if you want to work with ai, but that's just my opinion.

you don't need to dive deep into the details but it depends on what you want to do with it.

for the very least you need to learn the following things(everything is available on YT):
1. python(free code camp 10 hr long video)
2. little calculus (only differentiation)
3. data cleaning
4. basic statistics(optional)
5. feature engineering(essential)
6. learn about different machine learning models and their pros and cons also which perform better with which kind of data(there are enough libraries to provide abstraction)
7. some ml essentials, like distance, regularization, etc.
8. plotting of graphs.
9. hyperparameter tuning.
10. clustering algos with pros, cons, and what kind of data they work well with. (optional, you might not need clustering for your use case)
11. deep learning(optional but important, as you don't need deep learning if you are good with feature engineering).
12. Understand that your model will perform based on the quality of data(Garbage In Garbage Out), simple algorithms can work wonders with good data whereas state of the art algorithms would suffer from poor quality of data)

breaking this down still, I don't see how this could be explained to a non-technical person(I think I can but I am not getting how the ELI5 version will give the ability to a nontechnical person to use ai in business).

also not that as of now AI is not a one-stop solution that does everything, it's better to say it automates the ONE feature. eg: GPT3 only predicts next word based on the previous input.




most guys on this thread are more experienced than me and you are welcome to correct me or expand on what I wrote.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[Zealander]

Speaking as a software engineer, i am quite pessimistic about the people here who don’t know how to code and want to create an AI business. Don’t want to be a downer but there are a lot of people that work with AI and know it in and out, know what it can and can’t do and even they can’t create very profitable businesses. What makes you think that you can come in with no experience and create so much value there?

I think only thing that one can hope as a solopreneur/small operation is to just use AI in some clever niche to make it better and there is no need to try to create something like making AI do some grand stuff that it can’t do yet. There are teams of top scientists working in many areas already.

 

[The-J]

Report: Top 25 Use Cases for Marketing Artificial Intelligence

We asked hundreds of marketing professionals to rate the value of intelligently automating more than 60 uses cases. Here are their top 25.

www.marketingaiinstitute.com

Here are the top 25 marketing AI use cases, with average rating, according to our analysis of 210 respondents. As you can see, the highest rated use cases were scored between 3 and 4, indicating they are of moderate-to-high value for marketers.

• (3.88) Analyze existing online content for gaps and opportunities.
• (3.72) Choose keywords and topic clusters for content optimization.
• (3.71) Construct buyer personas based on needs, goals, intent and behavior.
• (3.70) Create data-driven content.
• (3.64) Discover insights into top-performing content and campaigns.
• (3.64) Measure return on investment (ROI) by channel, campaign and overall.
• (3.64) Adapt audience targeting based on behavior and lookalike analysis.
• (3.55) Optimize website content for search engines.
• (3.52) Recommend highly targeted content to users in real-time.
• (3.47) Assess and evolve creative (e.g. landing pages, email, CTAs) with A/B testing.
• (3.44) Deliver individualized content experiences across channels.
• (3.43) Define topics and titles for content marketing editorial calendars.
• (3.41) Predict content performance before deployment.
• (3.40) Forecast campaign results based on predictive analysis.
• (3.37) Build media and influencer databases based on interests, audiences and intent.
• (3.30) Prescribe strategies and tactics to achieve goals.
• (3.27) Present individualized experiences on the web and/or in-app.
• (3.26) Design websites, landing pages and calls-to-action.
• (3.26) Map buyer journey stages based on historical lead and conversion data.
• (3.24) Draft social media updates with copy, hashtags, links and images.
• (3.24) Analyze and edit content for grammar, sentiment, tone and style.
• (3.23) Determine goals based on historical data and forecasted performance.
• (3.23) Score leads based on conversion probabilities.
• (3.23) Curate content from multiple sources.
• (3.22) Build dynamic charts and graphs to visualize performance data.

 

[Alfie321]

I'm personally learning tensorflow as their JS implementation seems promising for the professional publishing niche with open source pre trained models. I work for a company who specializes in providing great building experiences for professional publishers so I do see a lot of potential thanks to INSIDERS advantage, however I do agree that in its current state, people think of AI like monkeys seeing fire. Most go by what Elon Musk thinks it is, or what they read in a buzz article. Honestly I think AI is a big bubble. Nobody knows (or wants to admit) they are doing statistics at scale and it does not has to do with intelligence at all.

For example, I remember a clickbait article saying facebook shut down their chat AI because it was too dangerous as they were "communicating with sentences we couldn't understand" when in fact they shut it down because it was crappy.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[Antifragile]

AI content platform Jasper raises $125M at a $1.5B valuation | TechCrunch

Jasper, a platform leveraging AI to generate marketing art and copy, has raised $125 million at a $1.5 billion valuation.

techcrunch.com

F*cking hell. Talk about having a Fastlane business. They launched in January 2021.

As you know, I have an annual subscription and use it daily.

Turns out, it’s helping me write most of my emails (important ones, only!). I take text that I write, copy/paste into Jasper and then click “re-phrase”. Boom … new ways of saying the same but better. Way better.

I can’t believe that’s how I ended up using it and even recently renewed the subscription for another year. It is expensive! And worth it for me. Love it.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[Empires]

I am not currently in the AI space but have plans for it in the future. The sheer amount of opportunity in the space is incredible. Will surely change an insane amount of industries over time.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[ChrisV]

@ChrisV
I bought the course you recommended, Analytics Vidhya. Why do they use Jupyter? I've never used an IDE that needed to be accessed from the terminal, so I found it a little strange.

Hm, not all versions need to be accessed from the terminal. If you're more comfortable you can launch it from Anaconda, or your web browser.

Honestly though, the Terminal is sooooooo useful when coding, and you might find yourself preferring it. But whatever you prefer for now. But I don't believe that the terminal is a requirement for running Jupyter.

 

[spreng]

I'm interested in seeing what developments AI makes, and maybe getting into a vertical aligned with it. Currently run 2 domains meant to generate b2b AI SaaS leads, will see how that goes . Big problem I've found is predicting buyer intent with b2b products, especially AI. Perhaps product reviews or specific white paper detailing a specific problem are the way to go.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.