Technical Interview, Machine Learning, Blockchain - Uniqtech Newsletter 2022 Q2
A reminder : by using our contents, reading our contents, and or using our sites and channels including but not limited to ml.learn-to-code.co, uniqtech.medium.com, uniqtech.substack.com, you agree with our terms of services, privacy policy, disclaimers and other legal terms. If you disagree, please immediately stop using all our services. More information can be found here:
The documents can be found here: https://ml.learn-to-code.co/payments.html Please review.
====================================================================================================
Tech power (Uniqtech Guide) for your technical interviews and leveling up developer skills.
====================================================================================================
We have gotten user feedback that our contents are flashcards are amazing yet our UI looks… terrible. Yes, we are aware. Our sincere apologies. Better UI and features are definitely our goal later this year. That’s our №1 priority to work on in the next YC startups school build sprint. Right now, we are just adding as many flash cards as possible, so we can launch courses and projects (all included in your subscriptions, coming soon). Developer NFT schedule is delayed to the dismal economic activities in the crypto world. Message or send us your feedback. You can also submit feedback by voting for flash cards. Each flash card has a vote button.
====================================================================================================
Easter eggs, BEST RESOURCES, TIPS
This month’s Easter egg (high quality resource, pro tip) for data science: Data Science and Machine Learning on Google Cloud Platform. Become a member today to access member only flash cards, easter eggs, pro tips.
https://ml.learn-to-code.co/payments.html
Pay via Stripe, Substack, or crypto.
Data Science and Machine Learning on Google Cloud Platform. Part 01 [pro]
https://ml.learn-to-code.co/skillView.html?skill=8wpEtemeSJUhMGcMNVgU
Data Science and Machine Learning on Google Cloud Platform. Part 02 [pro]
https://ml.learn-to-code.co/skillView.html?skill=YbxHTenyguMR90Wz6P3J
Data Science and Machine Learning on Google Cloud Platform. Part 03 [pro]
https://ml.learn-to-code.co/skillView.html?skill=HWDRMw5ukInQQbdkd03b
Super pro tip, easter egg. Best resource found on the internet :
Linear Algebra (tensor math) for developers. Linear Algebra for Developers [pro, advanced]
https://ml.learn-to-code.co/skillView.html?skill=2qjDBydFAHgAwudcvPhJ
====================================================================================================
Past newsletters are published here:
And here:
https://medium.com/data-science-bootcamp/newsletters/machine-learning-bytes
====================================================================================================
CAREER, TECHNICAL INTERVIEWS, JOBS
As the economy enters 💩state, we will be sure to send out more interview preps in the upcoming days so that you can get ready for cheap. Stay tuned. Reach out if you have hardships.
Goodbye Startups, Hello Big Boring Corps
Everyone Wants to Work for Big, Boring Companies Again — Analysis by Conor Sen | Bloomberg
Crypto winters and tech layoffs are no fun. Coinbase initially aimed to make 6000 hires then rescinded offers and cut off job postings except mission critical roles like security. What’s a better time to beef up your portfolio, redo your resume, and prep for technical interview? We may all need those data structure and algorithms skills again.
Why efficient algorithms (measured with Big O) really matter for big techs like Google, Meta (Facebook) and FANG companies? [pro]
https://ml.learn-to-code.co/skillView.html?skill=Ifo5QUa9xEmQuwTNrDwE
Fibonacci Sequence Big O [pro]
https://ml.learn-to-code.co/skillView.html?skill=tckym7y6VMCgUAnDAjxl
Relational Database Basics [public] primary key, foreign key
https://ml.learn-to-code.co/skillView.html?skill=usJr7a66CewCyEG36EYZ
We have said it many times: according to our analysis of job postings, successful tech companies (even FAAANG companies, big techs) still ask for SQL skills. What sets modern data engineers apart from previous generations of SQL developers is the ability to execute high performance SQL queries and handle big datasets. This could be attributed to infrastructure, indexing and technology change. For example, Google BigQuery allows data analysts to query Terabytes of data quickly, writing the good old SQL, no optimization needed on the user’s side. Job posts ask for intermediate and advanced SQL skills : nested queries, ETL, aggregation, complex JOINs.
Relational Databases use SQL or SQL like query languages. SQL JOINs statements are used to join relational data tables. JOINs are slow. They are useful because relational databases are all about relations and associations. We use JOINs to query different tables. Traditionally they have performance issues and can be very slow. Back then SQL queries can take hours, even a day. In modern data warehousing, e.g. BigQuery, server less, in the cloud, we can use SQL to query data, but the BigQuery infrastructure and optimization is so impressive, the speed and result is night and day compared to MySQL database. Analysts can query terabytes of data in minutes.
Hip tech companies often ask for nosql database skills. This is common for large gaming companies with millions of transactions and events. NoSQL database : MongoDB, firebase firestore. Alternatives to relational database and nosql databases: graph database Neo4j.
Databases, SQL, nosql, graph database [public]
https://ml.learn-to-code.co/skillView.html?skill=xAkoB4Z2QVS5dgckMemE
SQL datatypes [public]
https://ml.learn-to-code.co/skillView.html?skill=pWbUhuB3kUIuqw4xOeYr
Graph Algorithm use cases [pro]
https://ml.learn-to-code.co/skillView.html?skill=0pp7loAB5oZ0PHjndzWH
Devtools, IDE
Why use Pycharm? Great for python developers. Pycharm works well for large/ professional projects. Functionalities include: auto formatting, refractor, see changes. Surround with using ctrl + alt + T with express. Pycharm debugger, at the bottom of the screen or use short cut hot keys : Ctrl + D or Command + D. You can step through the code. Can stop at break points.
Why use VSCode? It’s popular among developers. It’s enterprise level software, developed by Microsoft. Why VS Code? Anecdotally, developers we have met often say they use Microsoft Visual Studio Code. StackOverflow developer survey confirmed that VS Code is the most popular IDE. 71.06% of professional developers choose VS Code. By the way we will discuss in the near future how VSCode is very useful for smart contract development in Solidity as well!
Devtools, IDE, pro tips for developers [public]
https://ml.learn-to-code.co/skillView.html?skill=cZ3lBg9aBsgFiSJR1CYs
Devtools, IDE. What do professional developers do? [pro, high quality]
https://ml.learn-to-code.co/skillView.html?skill=LXLHL85JrKBxQUuClkwv
Insights
Tech jobs at Quora, Q&A site
https://ml.learn-to-code.co/skillView.html?skill=oSanLAYFjK1VIcIpBGMx
Hashmap, Dictionary (Data Structure) [public]
https://ml.learn-to-code.co/skillView.html?skill=sTyiVBx5UwtFksq5GRBV
Pro tips & tricks : using % modulo operand and // integer division in interviews [pro]
https://ml.learn-to-code.co/skillView.html?skill=o6GhquXaNGkUnXtxAPYE
====================================================================================================
MACHINE LEARNING, DEEP LEARNING, AI
Cost Functions, Metrics,
Cost function [public]
https://ml.learn-to-code.co/skillView.html?skill=EG1YVO7Vi2saLHJK7qWK
Accuracy (definition) [public]
https://ml.learn-to-code.co/skillView.html?skill=4P0j8OIkXHk1MSTGzopM
Optimizers
Optimizer, Optimization algorithm [public]
https://ml.learn-to-code.co/skillView.html?skill=CqD7aJmSOSrp61S0fvvu
Optimization (definition) [public]
https://ml.learn-to-code.co/skillView.html?skill=VRWSgQPNl0KCVIJOskbc
What’s back propagation (BP)? [public]
https://ml.learn-to-code.co/skillView.html?skill=r09LuM3mHzfCCwgjSGmR
tensorflow optimizers
https://ml.learn-to-code.co/skillView.html?skill=BAv6pD1HJMk1vNcs5azP
Advanced machine learning
Simulated Annealing [advanced, pro]
https://ml.learn-to-code.co/skillView.html?skill=aOLhgHQMDLrRwRHrzGTw
What’s knowledge distillation? (Definition) [public]
https://ml.learn-to-code.co/skillView.html?skill=O7rxlxantpKasSyyYL9z
SpaCy NLP natural language processing timeline
https://ml.learn-to-code.co/skillView.html?skill=kI5442yZiNy0cZjYbO9a
How to get the most out of Kaggle #datascience competition [pro]
https://ml.learn-to-code.co/skillView.html?skill=lqjieIWC5ybtTlFj37Zq
Talking about advanced topics: did you know OpenAI has something fancy again called Dalle2. This time it can generate some fancy photos.
https://openai.com/dall-e-2/ Google quickly came up with an insanely realistic, yet imaginative, ground making text to image generator called Imagen. Previously we wrote about basics of OpenAI and its training methods here https://ml.learn-to-code.co/path3.html?tutorial=gpt3_partial
What is LightGBM (definition)? [public]
https://ml.learn-to-code.co/skillView.html?skill=LAwTEwZeQDrvk0eG2pK2
Using LightGBM with other machine learning models [pro, advanced] Advanced: how to use LightGBM as a part of a machine learning pipeline? [pro, advanced]
https://ml.learn-to-code.co/skillView.html?skill=7zpYyRM2zCCbgs5Dj6lo
====================================================================================================
Blockchain, Web3, NFT, Metaverse, Crypto
This year cryptos really have crashed and burned. While financially, being a web3 developer is no longer as appealing as it is 3 months ago, being a web3 developer, a builder, a creator is still a must have skill for the near future. Blockchains are here to stay as public databases. NFT transactions has reached high volumes in Ethereum, Solana and Tezos. Ivan from Moralis said Bull markets are for investors, Bear markets are for true builders.
set up smart contract developer tool [pro]
https://ml.learn-to-code.co/skillView.html?skill=y9QjadB4bOMnHfg6ha2L
Proof of Stake (PoS) explained by coinbase (definition) [public]
https://ml.learn-to-code.co/skillView.html?skill=dPTLF9GgTLopyIvN9Q1y
As data scientists and engineers, we can also discuss crypto analytics, it is interesting. An example question during the bear market: how many ethereum wallet addresses are still in profit?
====================================================================================================
MACHINE LEARNING TRIVIA
Why did Tensorflow 1.0 started as in graph mode rather than eager mode? [public, history of deep learning]
https://ml.learn-to-code.co/skillView.html?skill=YPXxiAgeUEKzEJHpL4Fq
#tensorflow #AI #graph #computegraph #deeplearning #AI
If you want to read more about the history of Tensorflow check out our Medium post.
Topic : compressing, optimizing machine learning models for edge computing. [public]
https://ml.learn-to-code.co/skillView.html?skill=N22PT64zbdXYSEK0xbrG
Machine learning Vs Traditional Data Science [public]
https://ml.learn-to-code.co/skillView.html?skill=99vCLEpExoQDaT7lwxac
====================================================================================================
We publish exclusively on our own sites only : ml.learn-to-code.co, and uniqtech.medium.com, and uniqtech.substack.com. Our contents should not be posted, scraped or shared elsewhere in its original form. Any educational institute can contact us to request copies, potentially free of charge for your students. White label service potentially available. Warning : we have not hired additional employees or representatives, we do not partner with any other company. Be careful of scams, or any one impersonating us. We do not have a presence outside the United States. We don’t have any deals or partnerships. Please only contact hi @ uniqtech .co . All other emails and promotions, contacts, business requests are false. We are not responsible for any external costs, damages, or requests. We do not currently offer any services, advisories, nor consulting. We have had a few Stanford interns in the past. We are happy to verify their identity and provide reference.