SDE & AI @Amazon | Simplifying software engineering, system design, and generative AI | Building Hungry Minds: the newsletter to stay ahead in software 👇

Tell me about the last time you felt growing as a software engineer. Really.

𝙽𝚞𝚕𝚕𝙿𝚘𝚒𝚗𝚝𝚎𝚛𝙴𝚡𝚌𝚎𝚙𝚝𝚒𝚘𝚗? Here's why:

We all face information overload as software engineers. 
There's too much content and too little time.
You have to stay relevant, but it's hard.

Right now, you're either working or taking a break.

You are hoping to get a dopamine shot from scrolling your feed.

You can find:

💡 A practical tip to implement
📚 An inspirational quote
😆 Something funny

And then what?

→ Scroll past it? 
→ Forward it blindly to a friend?
→ Forget about it in the bookmarks?

The harsh truth: nothing.

What you consume passively doesn't make you learn.

How do I know?
It was the same for me for years.

How do you stay updated with AI, stay relevant in the job market, and continuously grow?

That's the problem I set out to solve with Hungry Minds - a free weekly newsletter to help software engineers learn and stay updated, without the noise.

I'm doing that by:

1. Writing to you daily on LinkedIn to 40K+ followers and over 5M impressions last year
2. Compressing 1 week of software deep dives and trends in 1 free newsletter issue
3. Sharing my thoughts and learnings being promoted to SDE II in 1.5y at Amazon

If that sounds interesting, check out Hungry Minds or DM me! I'm always down for a chat.

Save 50+ hours of your week here: 𝗵𝘁𝘁𝗽𝘀://𝗵𝘂𝗻𝗴𝗿𝘆𝗺𝗶𝗻𝗱𝘀.𝗱𝗲𝘃 (6K+ engineers from startups to big tech are reading already)

DISCLAIMER: Opinions expressed are solely my own and do not express the views or opinions of my employer.

My career background:

I'm currently an L5 software engineer at Amazon in Madrid, working on core Search CX and Personalization. 
In 2022/2023, I delivered 2 successful billion-scale projects with multiple teams across different time zones.

I have worked with multiple code languages (although this doesn't make a difference), and with tech stack in the lines of AWS CDK, React applications, scheduling systems, backends with GraphQL, and simple AI applications.

My education background:

- I finished a Master's in Machine Learning and AI with l'Ecole Polytechnique and IPP
- I completed a Master's in Computer Science and Data Science with ESILV Paris
- I got a Master's in business development and marketing with EMLV Paris

Alexandre Zajac

𝗖𝗼𝗱𝗲 𝗱𝗼𝗰𝘂𝗺𝗲𝗻𝘁𝗮𝘁𝗶𝗼𝗻 𝗶𝘀 𝗡𝗢𝗧 𝗔𝗯𝗼𝘂𝘁:


→ One senior gate-keeping all the knowledge. 

Code documentation is ALL about:

→ Sharing knowledge so the whole team knows how the system works.

1-person docs = single point of failure.

Whole-team docs = redundancy + knowledge multiplication.

Remember:
You NEED documentation to manage a system through people and time.


How do you manage docs effectively?


P.S: Documente my personal interview cheatsheet: https://lnkd.in/d_eCGEJR

SDE & AI @Amazon | Building Hungry Minds to 1M+ | Daily Posts on Software Engineering, System Design, and AI ⚡

No alternative text description for this image

10+1 System Design Patterns Every Engineer Should Know in 2026:


___
btw, my personal interview cheatsheet: https://lnkd.in/d_eCGEJR
___


0.💡Load Balancer: 

Distributes incoming traffic across multiple servers to ensure no single server gets overwhelmed.

➡️ Useful when: You're scaling web applications or ensuring high availability.

1.💡CDN (Content Delivery Network):

Caches content at edge locations worldwide, reducing latency by serving users from the nearest node.

➡️ Useful when: You're serving static assets globally or streaming media at scale.

2.💡Message Queue:

Decouples services by letting them communicate asynchronously through a queue (like RabbitMQ or Kafka).

➡️ Useful when: You need reliable communication between microservices or handling traffic spikes.

3.💡Rate Limiter:

Controls the number of requests a user or service can make within a time window.

➡️ Useful when: You're protecting APIs from abuse or managing resource allocation.

4.💡Circuit Breaker:

Prevents cascading failures by stopping requests to a failing service and allowing it time to recover.

➡️ Useful when: You're building resilient microservices that depend on external APIs.

5.💡Database Sharding:

Splits a database into smaller pieces (shards) distributed across multiple servers.

➡️ Useful when: Your database is too large for a single server or you need horizontal scaling.

6.💡Caching Layer:

Stores frequently accessed data in memory (Redis, Memcached) to reduce database load.

➡️ Useful when: You're optimizing read-heavy applications or reducing latency.

7.💡API Gateway:

A single entry point that handles routing, authentication, and rate limiting for all your microservices.

➡️ Useful when: You're managing multiple services and need centralized control.

8.💡Reverse Proxy:

Sits in front of servers and forwards client requests, adding security, caching, and load balancing.

➡️ Useful when: You need SSL termination, compression, or hiding server architecture.

9.💡Event-Driven Architecture:

Services react to events rather than direct calls, enabling loose coupling and real-time processing.

➡️ Useful when: Building real-time analytics, notifications, or audit logging systems.

10.💡CQRS:

Separates read and write operations into different models, optimizing each for its specific purpose.

➡️ Useful when: You have complex domains with different read/write performance needs.


What did I miss?

Solving system design problems is HARD.

1 mega-template to solve any problem:


Approaching system design problems can be harder than coding problems.

For one reason: ambiguity.

13 concepts to use when solving a system design problem:


0. Load Balancer:

A load balancer efficiently distributes incoming network traffic across multiple servers to ensure no single server bears too much load.

1. API Gateway:

An API Gateway acts as a single entry point for all client requests, providing a unified interface to various microservices.

2. Cache:

Caching stores recently accessed data for speedy retrieval, reducing the load on databases and improving response times.

3. CDN:

A Content Delivery Network (CDN) stores static content closer to users to minimize latency and speed up content delivery.

4. Distributed File Storage:

Distributed file storage ensures data is redundantly stored across multiple locations for fault tolerance and easy access.

5. Search Index:

Using a search index like Elasticsearch allows for scalable and fast search capabilities across large datasets.

6. Notification Service:

A notification service handles the delivery of messages or alerts to users across different platforms and devices.

7. Data Processing Systems:

Data processing systems like Hadoop and Spark analyze large volumes of data, enabling scalable and complex computational operations.

8. Distributed Scheduler: 

A distributed scheduler orchestrates and manages the execution of distributed tasks across various computing resources.

9. Metadata Server:

The metadata server stores information about the structure of data, enabling efficient access and management.

10. Block Server:

Block servers manage the storage blocks in distributed file systems, which helps in data replication and retrieval.

11. Distributed Logging:

Distributed logging collects logs from various services in a centralized manner for monitoring and troubleshooting.

12. Replication and Partitioning:

Replication ensures data availability and durability, while partitioning helps in scaling databases by distributing data across different nodes.


Reference (aff.):

🔥 Grokking the system design interview: 
https://lnkd.in/dcMM8sZM


What did I miss?


~~~

👉🏻 Join 50,001+ software engineers getting curated system design deep dives, trends, and tools (it's free):
➔ https://lnkd.in/dkJiiBnf

~~~

If you found this valuable:

👨🏼‍💻 Follow Alexandre Zajac
🔖 Bookmark this post for later
♻️ Repost to help someone in your network

11 Algorithms Every Engineer Should Still Know in 2026:


0.💡 Breadth-First Search (BFS): 

Explore a graph level by level, starting from the root, which is great for finding the shortest path in unweighted graphs. 

➡️ Useful when: You're designing web crawlers or analyzing social networks.


1.💡 Two Heaps: 

Uses a min-heap and max-heap to manage dynamic datasets efficiently, maintaining median and priority. 

➡️ Useful when: You need to manage a priority queue or dynamic datasets.


2.💡 Two Pointers: 

This technique takes 2 points in a sequence and performs logic based on the problem.

➡️ Useful when: You are implementing sorting or searching functions.


3.💡 Sliding Window: 

Optimizes the computation by reusing the state from the previous subset of data. 

➡️ Useful when: You're handling network congestion or data compression.


4.💡 Depth-First Search (DFS): 

Explores each path to the end, ideal for situations that involve exploring all options like in puzzles. 

➡️ Useful when: You're working with graph structures or need to generate permutations.


5.💡 Topological Sort: 

Helps in scheduling tasks based on their dependencies. 

➡️ Useful when: You are determining execution order in project management or compiling algorithms.


6.💡 Merge Intervals: 

Optimizes overlapping intervals to minimize the number of intervals. 

➡️ Useful when: Scheduling resources or managing calendars.


7.💡 Backtracking: 

It explores all potential solutions systematically and is perfect for solving puzzles and optimization problems. 

➡️ Useful when: Solving complex logical puzzles or optimizing resource allocations.


8.💡 Trie (Prefix Tree): 

A tree-like structure that manages dynamic sets of strings efficiently, often used for searching. 

➡️ Useful when: Implementing spell-checkers or autocomplete systems.


9.💡 Flood Fill: 

It fills a contiguous area for features like the 'paint bucket' tool. 

➡️ Useful when: Working in graphics editors or game development.


10.💡 Segment Tree: 

Efficiently manages intervals or segments and is useful for storing information about intervals and querying over them. 

➡️ Useful when: Dealing with database range queries or statistical calculations.


What did I miss?

You are what you eat.

10 brain foods to grow as an engineer:


0. Why Clock Synchronization Makes Engineers Lose Sleep:
► https://lnkd.in/gkDbWARS

1. How ChatGPT Actually Stays on Track (by Neo):
► https://lnkd.in/ggNMHDuV

2. How to Build Internal AI Agents That Actually Work:
► https://lnkd.in/gkJKyQnm

3. Scaling LLMs to Larger Codebases:
► https://lnkd.in/gWwfibJQ

4. Connection Pooling Finally Explained in 6 Minutes (by Nikki):
► https://lnkd.in/gkx6eNSG

5. 5 Engineering Dogmas It's Time to Retire (by Anton):
► https://lnkd.in/gYTdTmap

6. The System That Eliminates Interruptions for 10x Engineers (by Francisco):
► https://lnkd.in/g65_i6jf

7. Why Logging Sucks (And Everyone Knows It):
► https://loggingsucks.com/

8. Internal Dev Tooling at Meta (from Gergely):
► https://lnkd.in/g_A_BxFd

9. The Future of Software Development Is Software Developers:
► https://lnkd.in/gvzpK2CM


I am packaging the other deep dives and trends in today's issue of Hungry Minds.

Join 50,001+ software engineers from big tech to startups here: https://lnkd.in/dkJiiBnf


What are you reading this week?


#softwareengineering #systemdesign #programming

https://strategizeyourcareer.com

I'm a Software Development Engineer at Amazon. 

1:1s with my managers were not enough for me to explore all the career topics I care about. So I decided to start writing.

Francisco Ramírez

The 10-second wait after submitting a prompt is the most dangerous part of your day.

It is too short for a break but long enough to kill your flow.

When the IDE says "Generating..." your brain craves a dopamine hit. You Alt-Tab to Slack.

I caught myself doing this on my personal laptop during Christmas.

I tried to check Slack like at work... but it wasn't even installed. It was just a muscle memory reflex.

You think you are multitasking. You are actually destroying your context.

Focus is like a plane. Taking off takes massive energy. Every time you switch tabs you have to land and take off again.

Francisco Manuel Soto Ramírez

SDE2 @Amazon | Author of 19,000+ subs Newsletter (strategizeyourcareer.com)

Urgency is often fake.

Some engineers rush because things feel urgent.

They don't have time to solve the hard problems

Productive engineers question deadlines.

They differentiate what's urgent and what's important.

Ship what really helps users, not what looks busy.

Standup update: “I answered tons of DMs yesterday”.

I don't think that looks good.

1. No visibility
2. This may not be your responsibility
3. Can't evaluate if the time spent is justified
4. Tons of distractions, no deep work

Instead, you can build a system to manage your communications.

Stop living in your DMs and start designing that system:


https://lnkd.in/efNQUTDU

Fixed identity is a problem for softare engineers.

Some engineers tie their value to typing code and reject AI or new tools.

They protect pride, not outcomes.

Productive engineers shift identity.

From code writer to system thinker.

Leverage beats effort. Results are the only thing that matters.

Own the system, not the syntax.

Unfinished work leaks mental memory.

Some developers leave tasks half done and carry bugs in their head all night.

Focus drops. Energy drops.

Productive engineers close loops.

End the day with a clear next step written down.

Free your brain to solve one hard problem at a time.

You are not a CPU running at full speed all day.

Productive engineers schedule hard problems in high energy windows, and leave low energy time for reviews and admin work.

Some engineers fight their energy and burn out.

Match work intensity to energy and you ship more with less stress

Two faces of the same problem:

Some engineers see feedback and think “I’m bad at this”, so they hide and waste hours stuck.

Others think “they’re wrong”, so they ignore feedback and stop learning.

Both cost time and team speed.

They are the same problem: A fixed mindset.

Productive engineers treat reviews as a shortcut.

Ask. Clarify. Use tools. Learn fast.

Pride is expensive.

Work expands like gas in a container.

Some engineers give tasks two weeks and fill every day of the second week polishing.

They ship late and call it "quality."

Productive engineers time box hard.

Define done. Cut scope. Ship.

Smaller containers force real progress.

Programming isn’t typing code.

It’s organizing complexity.

That’s the real job of a software engineer.

JSESSIONID

li_at

linkedin.com/in/fransotodev/recent-activity/all/

The productive engineer does not split work and life.

Some developers only build systems at work.
Then they forget about groceries and bills because they don't use a calendar in their real life.

Run the same systems at home. You have them at work for some reason.

Less noise. More focus for the things that matter in your life

You solve your hardest bugs away from the keyboard.

I used to feel guilty for walking away when I was stuck. I thought staring at the monitor forced a solution. I just got headaches.

Your brain needs to connect unrelated ideas. This happens when you relax. It is why the fix hits you in the shower or while commuting.

But you need a system to make this reliable instead of luck. It is called an anchor.

1. Write down the specific question you cannot answer.
2. Go for a walk or read a blog about a completely different tech stack.
3. Let your brain run in the background.

☝️ Find out here the productivity system to have better ideas:

https://lnkd.in/etdWKjbd

I stopped trusting my memory. I added safe points.

Before switching tasks, I write past, present, future in a scratchpad.

It is my working memory when interruptions hit.

Some engineers rely on recall. Productive engineers dump state and resume fast.

Find out the process in this article: 

https://lnkd.in/e_JeBZJd

Productive engineers move faster than others.

Not because they are smarter, but because they obsess over iteration speed.

They do not wait 12 minutes for a deployment to learn nothing.

They spin up a local server and test a change in under a minute.

Fast feedback beats perfect setups every time.

Stop trying to be the hero who answers every DM.

That's destroying your focus.

I felt important when I answered immediately, but I was actually a bottleneck for my team.

You are hiding important information in private chats.

Instead it's better to create a system that removes the need to have these many DMs

Current software engineers do something else while AI writes their code.

They would return and approve the changes, then submit a PR.

A teammate finds a bug in a logic branch.

They tell them. "My AI did that"

Everyone is mentally detaching from the problem.

Productive engineers keep their focus inside the code.

Some engineers switch tabs and lose their flow.

They'll feel drained after work because of these micro switches.

Better to stay present with the code when:
→ Solving complex logic.
→ Building new features.

And you can do something else only when:
→ Writing routine documentation.
→ Fixing simple typos.

Check the full system: 

https://lnkd.in/dfJs973k

Stop trying to build the next AWS from scratch.

Growing up, I thought great engineers were building everything again from scratch. I was wrong. That would just slow the product development.

Great engineers work in platforms, which are like coral reefs. They support use cases never imagined before.

- The creators of GPS tracked submarines. They never planned for Pokémon GO.
- Twitter did not invent the hashtag. Their users built it on top of the API.

You cannot compete by reinventing the wheel.

1. Build on top of existing cloud and libraries
2. Don't train AI models from scratch. Fine-tune existing checkpoints.
3. Write code that is a platform. Expose APIs so others can build what you did not predict.

That will also make the platforms stronger and more resilient.

Software engineers should get closer to other engineering fields.

Some software engineers think their problems are unique.

Then you watch civil or mechanical engineers ship atoms and steel.

Software is cheap to change.

That is an advantage. Use it.

Stop feeling guilty for browsing the internet.

I used to block every website to force myself to code. I thought focus was the only way to solve problems.

I was wrong. The web is a serendipity engine.

But it only works if you have an anchor.

An anchor is a specific problem you are holding in your mind.

When you have an anchor, your brain actively scans everything you see for a match.

You might read a random marketing blog and suddenly realize how to fix your distributed system.

You are not wasting time by reading blogs. You are widening your inputs.

Next time you are stuck try this:
1. Write your problem on a post-it note.
2. Find some time to relax and read some blogs and newsletters
3. Let your brain make connections.

Turn your relaxed time into brain background processing.

Slack interruptions are not small context switches. They are kill signals.

They force you to drop a half built thought with no save point.

You lose 20 minutes building context, then 20 more rebuilding it later.

Some engineers let their Slack set their priorities. Productive engineers design an attention architecture and decide what can interrupt them.

Every concept you learn is a LEGO piece.

Productive engineers know what each piece can do, so they reuse it to build faster. Some engineers chase new ideas, but progress comes from combining what you already know.

That is how you save time and ship smarter.

Keep learning.

taplio

The Battle

Alexandre Zajac

Francisco Ramírez

Followers

Posts

Engagements

Likes

Performance ScoreAlexandre Zajac's LinkedIn Performance

Performance ScoreFrancisco Ramírez's LinkedIn Performance

Ready to grow your LinkedIn presence?

taplio

The Battle

Alexandre Zajac

Francisco Ramírez

Followers

Posts

Engagements

Likes

Performance Score.css-goub6s{font-size:var(--chakra-fontSizes-md);font-weight:500;color:var(--chakra-colors-gray-600);display:block;margin-top:var(--chakra-space-1);}Alexandre Zajac's LinkedIn Performance

Performance ScoreFrancisco Ramírez's LinkedIn Performance

Ready to grow your LinkedIn presence?

Performance ScoreAlexandre Zajac's LinkedIn Performance