Thomas

💯 this is why I really like Learning mode in Claude Code I personally use this for all my side projects and it keeps me so much sharper, great if you want to use Claude Code but still stay hands-on! /config → Output style → Learning

@Hey_Eliana @grok what do you think?

You're building AI agents without a system. That's why they keep failing. Here’s the right system to go from idea → working agent 1. Define the job What problem are you solving? Who’s the user? What does success look like? 2. Design the brain Clear system prompt, role, instructions, guardrails (This is where most agents fail) 3. Pick the right model Speed vs cost vs intelligence Don’t overpay for simple tasks 4. Add tools APIs, databases, MCP servers, custom functions Agents become powerful when they can act, not just answer 5. Give it memory Short-term + long-term context So it learns, adapts, and improves over time 6. Orchestrate everything Workflows, triggers, retries, agent-to-agent communication 7. Build the interface Chat, app, API, Slack bot Make it usable, not just functional 8. Test + improve Evals, latency checks, real-world feedback Iteration is the real moat ❤️ Like 🔁 Retweet 🔖 Bookmark Follow @Hey_Eliana for more such posts #AI #ArtificialIntelligence #GenerativeAI #CareerGrowth #Upskilling

Less prompt-engineering. More systems.

If you want to become good at SYSTEM DESIGN (in 2026), learn these 18 case studies:

Spring Boot Advanced Annotations Cheatsheet ✅

Networking Commands and their Usage

𝗪𝗵𝘆 𝘆𝗼𝘂 𝗺𝘂𝘀𝘁 𝗿𝗲𝗮𝗱 "𝗗𝗲𝘀𝗶𝗴𝗻𝗶𝗻𝗴 𝗗𝗮𝘁𝗮-𝗜𝗻𝘁𝗲𝗻𝘀𝗶𝘃𝗲 𝗔𝗽𝗽𝗹𝗶𝗰𝗮𝘁𝗶𝗼𝗻𝘀" 𝗯𝗼𝗼𝗸 I read it twice, and here's what people who didn't read it are missing. After two decades in software, I thought I had a solid understanding of distributed systems. Then DDIA rewired how I feel about data. The book doesn't just explain technologies; it explains why they exist. Why does Cassandra use LSM-trees while PostgreSQL uses B-trees? DDIA shows you: LSM-trees optimize for write-heavy workloads through sequential writes, while B-trees optimize for reads through in-place updates. Simple trade-off, huge implications. Why does MongoDB default to single-leader replication instead of multi-master? Because write conflicts in multi-leader setups are brutal to resolve. The complexity rarely justifies the benefits. The real value isn't learning what these systems do. It's about learning how to reason about trade-offs: consistency vs. availability, latency vs. throughput, and simplicity vs. performance. Every architecture decision becomes clearer when you understand the fundamentals. Who should read it: 🔹 Mid-career engineers designing systems 🔹 Anyone preparing for systems design interviews 🔹 Architects who want to move beyond cargo-cult engineering Who might struggle: 🔸 New developers (assumes distributed systems background) 🔸 Anyone looking for quick tutorials (it's pure concepts) DDIA isn't light reading; it's 500 pages of dense material. But it's the difference between copying patterns and understanding principles. I wrote a complete breakdown covering what I loved, what I didn't, and the key takeaways that changed how I design systems. 👉 Read the complete analysis here: newsletter.techworld-with-milan.com/p/what-i-learn… Have you read DDIA? What was your biggest insight? #softwareengineering #programming #systemdesign

Salary Negotiation Cheatsheet

JOB INTERVIEW CHEAT SHEET

My recommended materials for cracking your next technical interview Coding - Leetcode - Cracking the coding interview book - Neetcode System Design Interview - System Design Interview Book 1, 2 by Alex Xu, Sahn Lam - Grokking the system design by Design Guru - Design Data-intensive Application book Behavioral interview - Tech Interview Handbook (Github repo) - A Life Engineered (YT) - STAR method (general method) OOD Interview - Interviewready - OOD by educative - Head First Design Patterns Book Mock interviews - Interviewingio - Pramp - Meetapro Apply for Jobs - Linkedin - Monster - Indeed Over to you: What is your favorite interview prep material? – Subscribe to our weekly newsletter to get a Free System Design PDF (158 pages): bit.ly/bbg-social

Popular Algorithms for System Design Interviews

№1 Mistake Everyone Ignores: Overdoing Soft Skills 📌 Here are 12 essential soft skills and how to find the golden mean to get the most out of them.

8 Soft Skills Techniques To Excel In Your Career.

4 GitHub Repositories to Prepare for 4 Different Types of Software Engineering Interviews: 1. System Design Interviews: github.com/ashishps1/awes… 2. Low Level Design Interviews: github.com/ashishps1/awes… 3. Coding Interviews: github.com/ashishps1/awes… 4. Behavioral Interviews: github.com/ashishps1/awes… ♻️ Repost to help others in your network

@Book_therapy223 @grok explain

How to take nothing personally

@mdancho84 @grok explain

All data scientists need to learn AI right now. Agents are the future of data science. This is why:

Another huge week in AI Agents 🧵  I summarized everything announced by AgentOps, Browser Use, LangChain, CoPilotKit, Triple Whale, Replit, Outshift by Cisco, Mastra AI, and more. Here's everything you need to know and how to make sense of it:  (save for later)

What is MCP? Why is everyone talking about it? Let’s take a closer look. Model Context Protocol (MCP) is a new system introduced by Anthropic to make AI models more powerful.

5 techniques to fine-tune LLMs, explained visually! Fine-tuning large language models traditionally involved adjusting billions of parameters, demanding significant computational power and resources. However, the development of some innovative methods have transformed this process. Here’s a snapshot of five cutting-edge techniques for finetuning LLMs, each explained visually for easy understanding. 1️⃣ LoRA: ↳ Introduce two low-rank matrices, A and B, to work alongside the weight matrix W. ↳ Adjust these matrices instead of the behemoth W, making updates manageable. 2️⃣ LoRA-FA (Frozen-A): ↳ Takes LoRA a step further by freezing matrix A. ↳ Only matrix B is tweaked, reducing the activation memory needed. 3️⃣ VeRA: ↳ All about efficiency: matrices A and B are fixed and shared across all layers. ↳ Focuses on tiny, trainable scaling vectors in each layer, making it super memory-friendly. 4️⃣ Delta-LoRA: ↳ A twist on LoRA: adds the difference (delta) between products of matrices A and B across training steps to the main weight matrix W. ↳ Offers a dynamic yet controlled approach to parameter updates. 5️⃣ LoRA+: - An optimized variant of LoRA where matrix B gets a higher learning rate. This tweak leads to faster and more effective learning. ______ Follow me → @akshay_pachaar ✔️ Everyday, I share insights and tutorials around ML, LLMs, RAG and AI Agents.

There is a collection of Real World use cases of ML and LLM Systems. There are more than 500 case studies here. What I like the most about this is that all of these are real world problems and this links to the company engineering blog detailing how they solved it. This is as real as it gets and actually gives you a sense of ML System design for real use cases. If you get a chance do spend sometime going through the cases of your interest. Let me know what your favourite read from here.