rikkon1

These 7 statistical analysis concepts have helped me as an AI Data Scientist. Let's go: 🧵

90% of AI Engineer interviews in 2026 are just these 12 concepts repeated:

@realmadrid @FCBayernES @10ardaguler @KMbappe - Vini did nothing this game expect that one assist & should have been subbed after 70th minute - referee??? absolutely not the best game for him. I actually believe he forgot camavinga had a yellow already & he looked like it - camavinga oh boy.

@theMadridZone Honestly they will experience a Joselu effect✨

How confident are you for tonight, Madridistas? 🧐

Michael Saylor just built a $1.2 billion yearly dividend bill on a business that makes no money. And it's getting bigger every month. Strategy has issued $11.3 billion of preferred stock across 5 different series since January 2025. The rates are brutal: 8% to 11.5%. Here is what he owes every year: - STRC: $731M at 11.5% - STRD: $135M at 10% - STRF: $128M at 10% - STRK: $112M at 8% - STRE: $89M at 10% Total: $1.2 billion per year. In cash. Forever. The software business? It lost $112 million in free cash flow last year. It cannot pay a single month of these dividends. So where does the money come from? Issuing new MSTR shares. Every dollar of dividends is funded by printing more common stock. Saylor has pre-funded the next 2 years with a $2.25 billion cash reserve raised from share issuance. But new preferred stock is being issued every single week. The reserve will run out. And there is still $30.5 billion of preferred capacity left. If he uses it, annual dividends could hit $4 billion per year. On April 12, Saylor said Bitcoin only needs to grow 2.05% per year to cover everything. The math is correct. But dividends are paid in cash. Bitcoin appreciation is not cash. To turn Bitcoin gains into cash, he has only 3 options: 1. Sell Bitcoin (he has promised never to do this) 2. Issue more MSTR shares (destroys shareholder value when stock trades below NAV) 3. Turn off dividends (STRF and STRE dividends escalate to 18% if unpaid) Right now MSTR trades at 0.79x its Bitcoin NAV. A 21% discount. This means every new share issued destroys Bitcoin per share. The flywheel that worked on the way up is now running in reverse. There is also $6.8 billion of convertible debt with holder put options exercisable by September 2028. If the stock stays low, bondholders can demand cash repayment. The reserve cannot cover both dividends and puts. The breakeven was 1.5% in February. It is 2.05% today. At current issuance pace it could be 3.4% by year end. The number Saylor presents as fixed is actually rising every month. Bitcoin does not appreciate at 2.05% per year in a straight line. It goes up 300% then crashes 77%. Dividends are due every month regardless of where Bitcoin is in the cycle. This is a timebomb. It does not need Bitcoin to fail. It only needs Bitcoin to be volatile on a timeline that meets $1.2 billion in annual cash obligations.

Mastercard Senior Data Engineer Latest Interview Questions which my student cracked Interview Rounds: 4 Rounds - 3 Technical and 1 Managerial Round For entry level role, there would be 1 coding round initially Tip: 90% questions are based on Job Description. So, prepare topics mentioned in Company’s Job Description very well. Focus Area: * SQL + Spark + System Design * Kafka with PySpark * Kafka with Databricks SQL Questions (Medium Level) * Explain Inner Join vs Outer Join vs Left Join with use case? * What do you understand by Self Join. Give example? * How do you count number of Credit vs Debit Transactions per user. What insight can you get from this? * Identify and flag a user who undergoes 3 transactions within 60 seconds? PySpark Questions * How do you optimize transformations in PySpark? * Difference between Repartition vs Coalesce * What happens internally while running a PySpark job? * How do you Debug a failed Spark job? Kafka Questions * Explain your real time experience in Kafka? * How do you tackle flaws? * How do you ensure high throughput and low latency in Kafka? * Explain designing of real time payment processing pipeline using Kafka? Kafka + Databricks Questions * Design a pipeline involving Kafka, Databricks and Delta Lake * How do you handle schema evolution from Kafka to Delta Lake. * What is major difference between Batch vs Streaming in Databricks. System Design Questions * Design a system which can handle millions of transactions every second. * How do you detect and avoid duplicate processing of transactions. Mastercard majorly focuses on real time transaction pipeline, avoiding data loss and fraud detection approach. If you have attended Mastercard Interview, then please mention in comments what questions you have faced.

The Generative AI Data Scientist career screener in 2026: ❌ You don’t need a PhD ❌ You don’t need months of math ❌ You don’t need “3 years experience” ✅ You need proof you can ship AI/DS projects: 1. One AI/ML project that solves a real business problem 2. Decision-ready output (not “here’s a chart”) 3. A workflow the business can rerun (not a one-off notebook) That’s what separates ML modelers from AI/DS System Builders. Want the pattern? I’m teaching it live (free): 👉 learn.business-science.io/registration-a…

Data engineering will start feeling like Microsoft excel soon enough! - data pipelines generated with Claude Code: companies threatened: Informatica, Talend, Fivetran - physical data modeling managed by Databricks liquid clustering Companies threatened: senior data engineers who pride themselves on data modeling, Snowflake - automated data analysis by heavy LLM models Companies threaten: Accenture, Tata Consultancy Services, Infosys, McKinsey & Company - visualization frontends simplified and more customized Companies threatened: Tableau, Salesforce, Preset - sprint management done with agents Companies threatened: Atlassian, Linear, Notion The transformation we’re seeing is really intense. I’m excited to see where things end up! What else do you see going away or changing a lot?

@MadridXtra Coach, vini, bellingham, mendy, diaz, camavinga, fran garcia, alaba, mastantuono, rudiger OUT. STAY -> valverde, carreras, tchounami, pitarch, guler, trent, rodrygo, mbappe, courtois. We need re build. NOW!!!

🚨 NEXT GAME: BAYERN (A)

@TestimonyBamid3 @chyngyozil We have too many players with same play style

@metuinverse @chyngyozil What's the reason for adding Bellingham now

This!!! We needed a goal, we had players in the box. Fucking hell!!! x.com/mmtv5004/statu…

@FabrizioRomano Coach, vini, bellingham, mendy, diaz, camavinga, fran garcia, alaba, mastantuono, rudiger OUT. STAY -> valverde, carreras, tchounami, pitarch, guler, trent, rodrygo, mbappe, courtois. We need re build. NOW!!!

🚨 Álvaro Arbeloa: “Wednesday will be our night against Bayern. We will give 200% of our energies”.

@Amr82_ Coach, vini, bellingham, mendy, diaz, camavinga, fran garcia, alaba, mastantuono, rudiger OUT. STAY -> valverde, carreras, tchounami, pitarch, guler, trent, rodrygo, mbappe, courtois. We need re build. NOW!!!

طلب الابقاء على مودريتش طلب التعاقد مع زوبيميندي اراد التعاقد مع شتيلر اراد قلب دفاع اضافي اراد وضع فيني على الدكه ورفض كل مطالبه ورغم ذالك قدم افضل بدايه بتاريخ النادي وتصدر بفارق مهول وظل منافس شرس بالدوري والابطال والكاس والنتيجه اقالته بعدها الخروج من جميع البطولات رسميا😂

@heis_fede Coach, vini, bellingham, mendy, diaz, camavinga, fran garcia, alaba, mastantuono, rudiger OUT. STAY -> valverde, carreras, tchounami, pitarch, guler, trent, rodrygo, mbappe, courtois. We need re build. NOW!!!

Florentino Perez’s mistakes in the last 2 transfer windows: •Signed Kylian Mbappe •Signed no replacement for Kroos •Signed no replacement for Modric • Kept Dani Ceballos in the team •Rejected Joshua Kimmich •Rejected Bruno Fernandes •Rejected Reuben Neves •Rejected Viktor Gyokeres •Rejected Kyle Walker •Rejected Virgil Van Dijk •Bottled Zubimendi signing •Bottled Leny Yoro signing •Loaned Endrick •Signed Mastantuono instead of Cherki/Olise OLD MAN IT IS TIME TO RETIRE

@AlberrtoRM2 Coach, vini, bellingham, mendy, diaz, camavinga, fran garcia, alaba, mastantuono, rudiger OUT. STAY -> valverde, carreras, tchounami, pitarch, guler, trent, rodrygo, mbappe, courtois. We need re build. NOW!!!

Soy muy viuda de Xabi Alonso y no me pienso esconder, creo que se precipitaron echándole. Esta plantilla no ha funcionado ni con Ancelotti, ni con Xabi, ni con Arbeloa, NO ES CULPA DEL ENTRENADOR. Hay que apuntar a los de arriba, no trabajan una mierda.

@TeamMbappeXtra Coach, vini, bellingham, mendy, diaz, camavinga, fran garcia, alaba, mastantuono, rudiger OUT. STAY -> valverde, carreras, tchounami, pitarch, guler, trent, rodrygo, mbappe, courtois. We need re build. NOW!!!

Real Madrid were 7 points ahead in Laliga until Vinicius Jr happened.

@UNIVERSO_1899 Coach, vini, bellingham, mendy, diaz, camavinga, fran garcia, alaba, mastantuono, rudiger OUT. STAY -> valverde, carreras, tchounami, pitarch, guler, trent, rodrygo, mbappe, courtois. We need re build. NOW!!!

Nunca me voy a olvidar de cómo celebraron el título en Octubre 😂

@FabrizioRomano Coach, vini, bellingham, mendy, diaz, camavinga, fran garcia, alaba, mastantuono, rudiger OUT. STAY -> valverde, carreras, tchounami, pitarch, guler, trent, rodrygo, mbappe, courtois. We need re build. NOW!!!

🚨 Álvaro Arbeloa when asked if La Liga is over: “We are Real Madrid. We will fight until the end”.

@MadridXtra Coach, vini, bellingham, mendy, diaz, camavinga, fran garcia, alaba, mastantuono, rudiger OUT. STAY -> valverde, carreras, tchounami, pitarch, guler, trent, rodrygo, mbappe, courtois. We need re build. NOW!!!

🚨 REAL MADRID ARE 3 GAMES WITHOUT A WIN.

@yfwKinz Coach, vini, bellingham, mendy, diaz, camavinga, fran garcia, alaba, mastantuono, rudiger OUT. STAY -> valverde, carreras, tchounami, pitarch, guler, trent, rodrygo, mbappe, courtois. We need re build. NOW!!!

this play from mbappe genuinely almost made me destroy my TV.