@panyang

stacking.working hard to reach FIRE🔥

[kompot @ethkompot]

A Bell Labs veteran gave a 45-minute lecture in 1995 on 1 question every ML engineer skips: why does mathematics work at all. Almost nobody watches it.

The man is Richard Hamming. He wrote the error-correcting code that keeps every hard drive, modem, satellite link, and QR scan from collapsing into noise.

His question sounds like nothing. Why do symbols on a board predict planets they were never built for. Why did complex numbers, invented as pure abstraction, turn out to run all of electrical engineering. Why does a matrix eigenvalue decide whether a bridge stands.

He calls it "the unreasonable effectiveness of mathematics." Newton wrote 1 equation and it held for the Moon, the tides, and a falling apple at once. Nobody ordered the universe to obey algebra.

Watch the middle. He argues we find math everywhere partly because our brains evolved to see it, then admits that answer leaves most of the mystery standing.

No slides. 1 old man, 1 chalkboard, 45 minutes.

You train models on linear algebra every day. He spent a career asking why it works.

[Anuj @anujcodes_21]

A man spends 50 years teaching at MIT.

He knows his time is running out.

So he records one last lecture — everything he knows, distilled into a single hour.

He died 5 months later.

This is that lecture.

The most important hour you'll watch this week.

Bookmark it for later!

[The Philippine Star @PhilippineStar]

‘AS IF BEING NOT MEMBERS OF THE BAR, WE WILL BE LESS JUDICIOUS IN RENDERING FAIR JUDGMENT’

Senator Panfilo Lacson appeared to take a swipe at a fellow senator-judge on the third day of the impeachment trial of Vice President Sara Duterte on Wednesday.

In a post on X, Lacson commended his colleagues for remaining composed despite what he described as repeated lectures from one member of the impeachment court.

"I commend my fellow senator-judges for keeping our cool and not engage a colleague who has been lecturing us since Day 1 of the impeachment trial, as if being not members of the bar, we will be less judicious in rendering fair judgment," he said. (X/Ping Lacson)

[Vivek Galatage @vivekgalatage]

A Gentle Introduction to Assembly Language Programming by Prof. Perry Kivolowitz

https://t.co/INuyUue2Qk

[Jesus Falcis 🇵🇭 @jesusfalcis]

🚨JUST IN: Robredo-Tulfo tandem beats Duterte-Marcos tandem in a March 2026 survey by OCTA Research group.

This is the first time that Leni Robredo beats Sara Duterte in a survey – but this one is with hypothetical tandems.

I strongly support a President Leni Robredo and Vice President Raffy Tulfo tandem.

Good Governance, Aksyon Agad, at Angat Buhay ang ihahatid sa LAHAT NG PILIPINO.

Much work needs to be done reaching out to voters from Mindanao and the poorest of the poor (Class E).

But this is a good sign that even if Sara Duterte survives the impeachment (which she won’t), she is very much beatable in 2028.

And wala pang ₱6.77 Billion Pesos reveal ng AMLC reports nuong March ah. What more if the survey was done now!

Don’t be disheartened by the Senate coup and Bato Dela Rosa na makapal ang mukha maging pugante.

We can’t win all battles. But we are winning the war.

And we shall work to win the war.

Puksain ang kasamaang sisira sa demokrasya at kultura ng Pamilyang Pilipino.

Puksain ang ta-traydor sa bansa at magbebenta sa atin sa Tsina, drug lords, and oligarchs.

Puksain at talunin ang Pamilyang Duterte sa pulitika. 👊

#Halalan2028

[Piyush Itankar @_streetdogg]

For anyone wanting to learn C, start here!

I created this course with one single goal - You should be able to very clearly reason about what the machine will do as a result of the C code.

The course starts off with a mental model of how the CPU-Memory interactions works, then we move to instruction encoding for the RISC-V architecture, next, I explain how CPU boots up and starts to execute the first line of code written in Assembly. Along the way, I demo all claims based on QEMU based RISC-V machine.

We then explore what are the preconditions for the compiled C code to work and how to make the CPU jump to that code. Once we are done with that, we start to explore the keywords and constructs in the C language and I prove every claim by a demo involving GDB, Disassembly and reasoning down to what the machine does.

More details are available here: https://t.co/YWGn1If8gB

Also, 65 students rated this course
-- 61 rated it 5 Stars
-- 4 rated it 4 star!

You will learn a lot more than Just C. You'll learn how the machine works and how to control it.

[trish @TrisH0x2A]

computer science from scratch(literally)
from semiconductor transistors to programming in c.

[Onur Mutlu @_onurmutlu_]

If you would like study how computers & computing paradigms (e.g., CPU, GPU, NPU/TPU/systolic arrays, SRAM, DRAM) work, lecture videos from Spring 2025 edition of our beginner course, Digital Design & Computer Architecture are all online:

https://t.co/tKmpuxvVIx

@SAFARI_ETH_CMU

[logic destroyer @splinedrive]

Every FPGA book I’ve seen is the same crap. Same smoke and mirrors, same polished nonsense people pretend is useful. A lot of this stuff gets written and sold because people want to buy the idea of learning, not because it actually makes them good.

The best place to start with FPGAs is https://t.co/krfVnOymC4. Work through that site properly and you’ll come out with strong fundamentals.

[Vivek Galatage @vivekgalatage]

An algorithm course to revisit.

https://t.co/0UiVaw8T4t

[Chaofan Shou @shoucccc]

Claude code source code has been leaked via a map file in their npm registry!

Code: https://t.co/jBiMoOzt8G

[SumitM @SumitM_X]

As a senior Java backend developer,

It will be good if you have an understanding of the below 40 topics👇

1. CAP Theorem
2. Consistency Models
3. Distributed System Architectures
4. Socket Programming (TCP/IP and UDP)
5. HTTP and RESTful APIs
6. Remote Procedure Call (RPC) - gRPC, Thrift, RMI
7. Message Queues (Kafka, RabbitMQ, JMS)
8. Java Concurrency (ExecutorService, Future, ForkJoinPool)
9. Thread Safety and Synchronization
10. Java Memory Model
11. Distributed Databases (Cassandra, MongoDB, HBase)
12. Data Sharding and Partitioning
13. Caching Mechanisms (Redis, Memcached, Ehcache)
14. Zookeeper for Distributed Coordination
15. Consensus Algorithms (Paxos, Raft)
16. Distributed Locks (Zookeeper, Redis)
17. Spring Boot and Spring Cloud for Microservices
18. Service Discovery (Consul, Eureka, Kubernetes)
19. API Gateways (Zuul, NGINX, Spring Cloud Gateway)
20. Inter-service Communication (REST, gRPC, Kafka)
21. Circuit Breakers and Retry Patterns (Hystrix, Resilience4j)
22. Load Balancing (NGINX, Kubernetes, Ribbon)
23. Failover Mechanisms
24. Distributed Transactions (2PC, Saga Pattern)
25. Logging and Distributed Tracing (ELK Stack, Jaeger, Zipkin)
26. Monitoring and Metrics (Prometheus, Grafana, Micrometer)
27. Alerting Systems
28. Authentication and Authorization (OAuth, JWT)
29. Encryption (SSL/TLS)
30. Rate Limiting and Throttling
31. Apache Kafka for Distributed Streaming
32. Apache Zookeeper for Coordination
33. In-memory Data Grids (Hazelcast, Infinispan)
34. Akka for Actor-based Concurrency
35. Event-Driven Architecture: Event sourcing and CQRS (Command Query Responsibility Segregation).
36. Cluster Management: Kubernetes for container orchestration.
37. Cloud-Native Development: Using cloud platforms (AWS, GCP, Azure), and serverless computing (e.g., AWS Lambda).
38. Distributed Data Processing: Frameworks like Apache Spark or Apache Flink for large-scale data processing.
39. GraphQL: Alternative to REST for inter-service communication.
40. JVM Tuning for Distributed Systems: Memory management and performance tuning in distributed environments.

[freeCodeCamp.org @freeCodeCamp]

OpenStreetMap is a free, open alternative to Google Maps you can use for many location-based apps.

In this tutorial, @joshuaaabraham explains how OpenStreetMap works and how to integrate it into a React app with Leaflet.

You'll also learn about geocoding with Nominatim, when to choose OpenStreetMap over Google Maps, and more.

https://t.co/l3UbxEa7jw

[Dhanian 🗯️ @e_opore]

𝗛𝗼𝘄 𝘁𝗼 𝗗𝗲𝘀𝗶𝗴𝗻 𝗮 𝗡𝗲𝘄𝘀 𝗙𝗲𝗲𝗱 𝗦𝘆𝘀𝘁𝗲𝗺 (𝗜𝗻𝘀𝘁𝗮𝗴𝗿𝗮𝗺)

Design a 𝗵𝗶𝗴𝗵𝗹𝘆 𝗽𝗲𝗿𝘀𝗼𝗻𝗮𝗹𝗶𝘇𝗲𝗱, 𝗹𝗼𝘄-𝗹𝗮𝘁𝗲𝗻𝗰𝘆 𝗻𝗲𝘄𝘀 𝗳𝗲𝗲𝗱 𝘀𝘆𝘀𝘁𝗲𝗺 that aggregates content from followed users and recommended sources, ranks it by relevance using machine learning, and delivers it in a paginated, infinitely scrolling format to millions of concurrent users.

The system operates on a 𝗽𝘂𝘀𝗵-𝗽𝘂𝗹𝗹 𝗵𝘆𝗯𝗿𝗶𝗱 𝗮𝗿𝗰𝗵𝗶𝘁𝗲𝗰𝘁𝘂𝗿𝗲 that balances write amplification with read latency. When a user posts content, it is "fanned out" (pushed) to the feeds of active followers via a Feed Write Service. For users with millions of followers (celebrities), the system switches to a pull-based model where the feed is generated on-demand, avoiding the massive write overhead.

When a user opens the app, a Feed Read Service retrieves their personalized feed from a distributed cache. The feed is a pre-computed list of post IDs, ranked by the Ranking Service based on factors like recency, engagement, relationship with the author, and machine learning signals. Thumbnails and media are then fetched from a CDN for fast rendering.

The platform's core consists of 𝗵𝗶𝗴𝗵𝗹𝘆 𝗼𝗽𝘁𝗶𝗺𝗶𝘇𝗲𝗱, 𝗲𝘃𝗲𝗻𝘁-𝗱𝗿𝗶𝘃𝗲𝗻 𝘀𝗲𝗿𝘃𝗶𝗰𝗲𝘀:
- Feed Write Service: Handles fan-out of new posts to follower feed caches.
- Feed Read Service: Retrieves and paginates the ranked feed for users.
- Ranking Service: The ML-powered brain that scores and orders posts for personalization.
- Post Storage Service: Manages post metadata, captions, and comments.
- Media Service: Handles image/video upload, processing, and CDN delivery.
- Timeline Cache**: Distributed in-memory store (Redis) holding the pre-computed feed for each user.

Behind the scenes, an 𝗮𝘀𝘆𝗻𝗰𝗵𝗿𝗼𝗻𝗼𝘂𝘀 𝗲𝘃𝗲𝗻𝘁-𝘀𝘁𝗿𝗲𝗮𝗺𝗶𝗻𝗴 𝗽𝗶𝗽𝗲𝗹𝗶𝗻𝗲 powers personalization. Every view, like, and follow generates an event consumed by the Ranking Service, which continuously updates user embeddings and refines scoring models offline. Batch jobs pre-compute recommendations for cold-start users and generate the "Explore" feed.

This scale demands 𝗮𝗴𝗴𝗿𝗲𝘀𝘀𝗶𝘃𝗲 𝗰𝗮𝗰𝗵𝗶𝗻𝗴 𝗮𝗻𝗱 𝗶𝗻𝘁𝗲𝗹𝗹𝗶𝗴𝗲𝗻𝘁 𝗳𝗮𝗻𝗼𝘂𝘁 𝘀𝘁𝗿𝗮𝘁𝗲𝗴𝗶𝗲𝘀. Active users have their feed pre-computed and cached. Inactive users have their feed generated on-demand. The fan-out process is sharded by follower ID and processed asynchronously via message queues to absorb write spikes.

𝗖𝗿𝗶𝘁𝗶𝗰𝗮𝗹 𝗗𝗲𝘀𝗶𝗴𝗻 𝗣𝗿𝗶𝗻𝗰𝗶𝗽𝗹𝗲𝘀: 𝟭) 𝗛𝘆𝗯𝗿𝗶𝗱 𝗣𝘂𝘀𝗵/𝗣𝘂𝗹𝗹 𝗔𝗿𝗰𝗵𝗶𝘁𝗲𝗰𝘁𝘂𝗿𝗲 to handle celebrity users, 𝟮) 𝗠𝗟-𝗣𝗼𝘄𝗲𝗿𝗲𝗱 𝗥𝗮𝗻𝗸𝗶𝗻𝗴 for personalization, 𝟯) 𝗠𝘂𝗹𝘁𝗶-𝗧𝗶𝗲𝗿 𝗖𝗮𝗰𝗵𝗶𝗻𝗴 (feed IDs → post metadata → media), 𝟰) 𝗔𝘀𝘆𝗻𝗰𝗵𝗿𝗼𝗻𝗼𝘂𝘀 𝗙𝗮𝗻𝗼𝘂𝘁 with queue-based processing, 𝟱) 𝗘𝘃𝗲𝗻𝘁-𝗗𝗿𝗶𝘃𝗲𝗻 𝗣𝗲𝗿𝘀𝗼𝗻𝗮𝗹𝗶𝘇𝗮𝘁𝗶𝗼𝗻 for continuous model updates.

𝗧𝗲𝗰𝗵𝗻𝗶𝗰𝗮𝗹 𝗔𝗿𝗰𝗵𝗶𝘁𝗲𝗰𝘁𝘂𝗿𝗲 𝗦𝘁𝗮𝗰𝗸:
- 𝗙𝗿𝗼𝗻𝘁𝗲𝗻𝗱: React, React Native, Swift (iOS), Kotlin (Android)
- 𝗕𝗮𝗰𝗸𝗲𝗻𝗱 𝗦𝗲𝗿𝘃𝗶𝗰𝗲𝘀: Python (Django), Java, Go, Node.js
- 𝗗𝗮𝘁𝗮𝗯𝗮𝘀𝗲𝘀: PostgreSQL, Cassandra, DynamoDB
- 𝗙𝗲𝗲𝗱 𝗖𝗮𝗰𝗵𝗲: Redis, Memcached (for pre-computed timeline IDs)
- 𝗠𝗲𝗱𝗶𝗮 𝗦𝘁𝗼𝗿𝗮𝗴𝗲: AWS S3 + Global CDN (CloudFront, Akamai)
- 𝗠𝗲𝘀𝘀𝗮𝗴𝗶𝗻𝗴 & 𝗦𝘁𝗿𝗲𝗮𝗺𝘀: Apache Kafka, AWS Kinesis
- 𝗠𝗮𝗰𝗵𝗶𝗻𝗲 𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴: PyTorch, TensorFlow (for ranking and recommendation models)
- 𝗢𝗿𝗰𝗵𝗲𝘀𝘁𝗿𝗮𝘁𝗶𝗼𝗻: Kubernetes, Docker, AWS EKS
- 𝗠𝗼𝗻𝗶𝘁𝗼𝗿𝗶𝗻𝗴: Prometheus, Grafana, ELK Stack

👉 Learn more in The Modern System Design Handbook: https://t.co/2LauJpfJ9C
👉 Grab the Master System Design Case Studies: https://t.co/ujlZXVdJPO

[Matt Dancho (Business Science) @mdancho84]

🚨BREAKING: Google just DROPPED a masterclass on GPUs

Get it here 100% free:

[Julian Derry @CyberSamuraiDev]

You don’t remember how many times you unlocked your phone today. Your phone does.

Every lock and unlock is logged with a timestamp, building a pattern of your behavior over time. This lives in a database called KnowledgeC.db

No spyware. No hacking. Just your normal device. Now imagine what else it’s keeping track of.

[tut_ml @tut_ml]

Most people treat Claude Code like a smarter chat window.

That works… until your project grows.

This structure highlights something deeper: once you move beyond single prompts, you need separation of concerns. The same principles we use in software engineering apply here, too.

Look at the layout carefully.

https://t.co/YF2dqGxlFx is not just a note file. It becomes project memory.

It defines:

→ Standards

→ Constraints

→ Tone

→ Non-negotiables

→ Guardrails

Instead of repeating instructions in every prompt, you centralize them. That reduces token waste and behavioral drift.

Then you see skills/.

This is where things get powerful. A skill is essentially a reusable workflow.

If you’re repeatedly doing:

-Code reviews

-Refactoring

-Output formatting

-Structured analysis

It should not live in an ad-hoc prompt. It should live as a reusable capability.

That shifts you from prompting to system design.

Next, hooks/.

Hooks are underrated. They let you enforce checks:

→ Clean tool output

→ Validate structure

→ Log commands

→ Transform JSON

If you’re not using hooks, you’re manually correcting outputs that could have been automated.

Then the repository itself stays modular:

-docs/ for architecture decisions

-src/ for actual logic

-tools/ for scripts and utilities

This prevents your AI layer from bleeding into your application layer.

When I started organizing projects this way, three things improved:

-Fewer repeated instructions

-More predictable outputs

-Easier collaboration

Especially once you add:

→ Subagents

→ MCP integrations

→ GitHub Actions automation

→ Plugin development

Without structure, context becomes clutter. With structure, Claude operates within clear boundaries.

This is not about making things complex. It’s about treating AI workflows like first-class engineering components instead of temporary chat experiments.

If you're learning Claude Code and want to see how I implement this step by step, from installation to CLI usage, skills, hooks, subagents, MCP, GitHub Actions, and plugins, I’ve recorded the full process while building real workflows.

This is the Claude Code Full Course Link- https://t.co/vyorOTkdVs

Image Credit- Brij Kishore Pandey

Happy Learning!

#ClaudeCode #claudeai

[PCIJ @PCIJdotOrg]

After PCIJ posted an explainer on pro-China disinformation, the Chinese Embassy accused it of being a U.S. tool. Minutes after the embassy’s late-night Facebook post, pro-Duterte influencers amplified it, mirroring the influence patterns documented in the report.

[PCIJ @PCIJdotOrg]

While most Filipinos were asleep, 107 accounts with nearly 3 million combined followers shared the Chinese Embassy’s post within two hours. By the time their audiences woke up, it had already penetrated the pro-Duterte ecosystem and surpassed 400 shares within 24 hours.

[The Philippine Star @PhilippineStar]

‘PATTERNS OF DRUG USE DO NOT EXIST IN A VACUUM’

Paolina Massidda, common legal representative of victims of former president Rodrigo Duterte’s drug war, highlighted the realities of why poor communities are vulnerable to drug use.

“In these communities, structural conditions including the absence of adequate mental health services, limited access to quality education, scarce employment opportunities, and the chronic stress associated with poverty, create an environment in which substance use is more likely to emerge,” she said during the confirmation of charges hearing on Tuesday.

WATCH: https://t.co/zjphUMKZc0