The hottest Software Engineering Substack posts right now

And their main takeaways
Category
Top Technology Topics

[in case you missed it] Data Science Weekly - Issue 387

Data Science Weekly Newsletter • 0 implied HN points • 25 Apr 21
🕹 Technology Software Engineering
  1. Goodreads lets users decide what counts as a classic book, showing how the definition has changed over time. This online platform helps readers share their thoughts in various ways.
  2. Scientists are trying to decode whale language using AI, aiming to understand how these marine animals communicate. This research could reveal insights about their behavior and society.
  3. New techniques allow neural networks to solve tough equations much faster. This improvement can help us better model complex systems, making it easier for researchers and engineers.

[in case you missed it] Data Science Weekly - Issue 379

Data Science Weekly Newsletter • 0 implied HN points • 28 Feb 21
🕹 Technology Software Engineering
  1. Writing a book about data science can be a fun way to share knowledge and inspire others. It's also possible to make money online while doing it.
  2. Understanding Python concurrency is important for data scientists. Learning about topics like async and threads can boost your software engineering skills.
  3. Feature stores are essential for operationalizing machine learning. They help teams manage and deploy machine learning features efficiently.

[in case you missed it] Data Science Weekly - Issue 366

Data Science Weekly Newsletter • 0 implied HN points • 29 Nov 20
🕹 Technology Software Engineering
  1. Pinterest improved its data infrastructure by moving from Lambda to Kappa architecture to better handle its visual signals for machine learning. This change aimed to streamline costs and enhance signal availability.
  2. When building machine learning models, companies like DoorDash face huge data challenges. Choosing the right feature store is crucial for managing this data effectively, ensuring performance without overspending.
  3. Differentially private learning still faces challenges in performance compared to traditional models. For effective results, more private data or improved features from public data may be necessary.

[in case you missed it] Data Science Weekly - Issue 362

Data Science Weekly Newsletter • 0 implied HN points • 01 Nov 20
🕹 Technology Software Engineering
  1. Using AI for form extraction can greatly help fields like journalism and medicine. This could be more impactful than just predictive models.
  2. Data intuition is an important skill for data scientists. It helps them avoid being misled by bad data and analyses.
  3. Data engineering and data science are interconnected, but they have different focuses. Data engineering deals with preparing data, while data science analyzes it for insights.

[in case you missed it] Data Science Weekly - Issue 358

Data Science Weekly Newsletter • 0 implied HN points • 04 Oct 20
🕹 Technology Software Engineering
  1. Data quality is really important for machine learning to work well. If the data is bad, it can mess up the whole project and make people doubt the results.
  2. The State of AI Report covers current trends and future predictions in artificial intelligence. It looks into research advances, talent availability, and the impact of AI on industries.
  3. Using mobile phone data can help understand and manage the COVID-19 pandemic. However, it's crucial to consider what types of behaviors and populations this data represents.
Get a weekly roundup of the best Substack posts, by hacker news affinity:

[in case you missed it] Data Science Weekly - Issue 355

Data Science Weekly Newsletter • 0 implied HN points • 13 Sep 20
🕹 Technology Software Engineering
  1. DeepMind and Google Maps teamed up to improve travel time predictions using advanced technology called Graph Neural Networks. This helps users get even more accurate arrival times in busy cities.
  2. AI technology is now being used to spot edited videos, like deepfakes, by detecting hidden signals called 'deepfake heartbeats'. This could make it easier to tell which video was made with what software.
  3. A new book aims to teach machine learning from scratch, breaking down complex algorithms to make them understandable. It's a good resource for anyone wanting to learn the basics of machine learning.

[in case you missed it] Data Science Weekly - Issue 342

Data Science Weekly Newsletter • 0 implied HN points • 14 Jun 20
🕹 Technology Software Engineering
  1. There hasn't been a significant recent change in job automation from 1999 to 2019 in the U.S., even with new technology. Many jobs haven't become more automated, and pay rates for these jobs haven't really changed either.
  2. OpenAI offers an API that anyone can use for various language tasks. It allows users to perform tasks like translation and sentiment analysis without needing much prior knowledge.
  3. Managing technical debt in machine learning is important because many new data scientists don't learn best practices. This can lead to messy code that is hard to put into production, wasting time and resources.

Grow Your Software Engineer Career Series

Tech Ramblings • 0 implied HN points • 27 Jul 24
🕹 Technology Software Engineering
  1. Focus on clear communication to improve your engineering skills. Being able to explain your ideas can help you stand out.
  2. Avoid over-engineering your solutions. Keep things simple and practical to meet the needs without complicating the process.
  3. Having a supportive work environment is important. Building good relationships with your coworkers can lead to a better work experience.

Mastering DataFrame Joins in Spark: A Comprehensive Guide with Examples

DataSketch’s Substack • 0 implied HN points • 23 Jul 24
🕹 Technology Software Engineering
  1. DataFrames in Spark are like tables for big data. They help people work with large datasets efficiently across different computers.
  2. There are several types of joins in Spark, such as inner, left, right, and full outer joins. Each type has a specific way of combining data from two DataFrames.
  3. Setting up Spark is easy. You can install it, write a few lines of code to create DataFrames, and start joining data for analysis.

Unlocking Insights with Apache Spark: A Beginner's Guide to Data Aggregation

DataSketch’s Substack • 0 implied HN points • 03 Apr 24
🕹 Technology Software Engineering
  1. Apache Spark is a powerful tool for analyzing big data due to its speed and user-friendly features. It helps data engineers to work with large datasets effectively.
  2. Data aggregation involves summarizing data to understand trends better. It includes basic techniques like summing and averaging, grouping data by categories, and performing calculations on subsets.
  3. Windowing functions in Spark allow for advanced calculations, like running totals and growth rates, by looking at data relative to specific rows. This helps to analyze trends without losing the detail in the data.

Keep It in the Controller

inelegant puzzles • 0 implied HN points • 18 Oct 24
🕹 Technology Software Engineering
  1. It's okay to keep some business logic in your controllers as long as things stay organized. This can make your code easier to understand.
  2. You don't always need to extract code right away. Sometimes, it's better to wait and see how often you really need that logic in other places.
  3. Be careful about making things too complex too soon. It's important to separate domain logic from HTTP requests but also stay flexible for future changes.

TTW Extra #5 🔥: Must-watch 2024 QCon talks

Tech Talks Weekly • 0 implied HN points • 04 Jun 24
🕹 Technology Software Engineering
  1. QCon talks cover a wide range of software engineering topics, including backend, frontend, AI, and DevOps. These talks are great for anyone looking to learn more about tech trends.
  2. A curated list of 35 must-watch talks from QCon London and San Francisco includes interesting topics like how Netflix uses Java and scaling with Amazon DynamoDB. These videos can help you understand real-world applications of technology.
  3. If you subscribe, you'll get a weekly email with new talks from over 100 conferences. This is an easy way to stay updated on tech without the clutter of YouTube.

3 styling systems in 1 React app

Andrew's Substack • 0 implied HN points • 14 Oct 24
🕹 Technology Software Engineering
  1. Use Tailwind for most styles in your React app since it helps maintain consistency and keeps stylesheets small.
  2. CSS modules are helpful for specific cases like custom animations or grid areas when Tailwind isn't enough.
  3. For very dynamic styles that depend on JavaScript, using inline styles in React is the way to go, but these situations should be rare.

#88

The Nibble • 0 implied HN points • 09 Dec 24
🕹 Technology Software Engineering
  1. Meta is planning to build a huge subsea cable to improve its data traffic capabilities around the world. This project would be quite large and expensive, but it's still in the early planning stages.
  2. OpenAI is launching updates over 12 days to share its latest advancements and features. It's a great way for them to keep the community informed about what's coming next.
  3. Vitalik Buterin has shared his thoughts on what a crypto wallet should include, highlighting the importance of security and privacy features. This is crucial for users who want to feel safe with their digital assets.

Low-code API Prototyping Using Webhooks

The API Changelog • 0 implied HN points • 28 Nov 24
🕹 Technology Software Engineering
  1. Webhooks are a great tool for creating API prototypes quickly and easily. They allow different applications to communicate without needing a lot of coding.
  2. Using webhooks, you can set up various actions like listing, retrieving, and creating items in an API. For example, you can manage a coffee list with simple GET and POST requests.
  3. This method is user-friendly and helps you get feedback quickly. It's perfect for testing ideas without a complex setup.

HCF EP 001: Empty Project Setup

Hasen Judi • 0 implied HN points • 10 Dec 24
🕹 Technology Software Engineering
  1. The project 'HandCraftedForum' is built using a custom mini framework that includes components for data storage, server-side logic, and client-side UI creation. It's designed to make programming straightforward with a focus on using data and procedures.
  2. The initial setup creates a basic application skeleton that allows for running a local web server. This setup serves a simple welcome message while ensuring easy communication between the client and server.
  3. The licensing approach for this project allows others to use and learn from the code, but prevents commercial use of the final product. This way, the creator can protect the product while still fostering education and community contribution.

Announcement: Hand Crafted Forum

Hasen Judi • 0 implied HN points • 10 Dec 24
🕹 Technology Software Engineering
  1. There is a lack of good forums available, so creating a new one can help fill that gap. It’s something many people are familiar with.
  2. The development of this forum will focus on straightforward programming, using simple data and functions without complex structures. This makes it easier for more people to understand and contribute.
  3. The project will grow over time, starting with basic features and improving the user experience gradually. The aim is to make it simple to set up and use from the beginning.

2025

domsteil • 0 implied HN points • 31 Dec 24
🕹 Technology Software Engineering
  1. AI is changing how we interact and build software. It allows developers to create programs much faster and more efficiently than before.
  2. New AI technologies are making it easier for everyone to access and utilize smart systems in their daily tasks, potentially leading to a big shift in how businesses operate.
  3. In the future, software development will focus on using AI to handle tasks automatically. This will not only change how software is built but also how we measure success and pricing in business.

Why Aren’t APIs Using Code on Demand?

The API Changelog • 0 implied HN points • 17 Jan 25
🕹 Technology Software Engineering
  1. APIs could improve user experience by using code-on-demand, similar to how web browsers run JavaScript. This would allow APIs to deliver more interactive and efficient services.
  2. The lack of a standardized client for REST APIs makes implementing code-on-demand challenging. New formats like HyperMap are being developed to help change that.
  3. Concerns about security hold back the adoption of code-on-demand in APIs, but the potential benefits suggest it could lead to better features and functionality.

Selectively Serving Your API Reference

The API Changelog • 0 implied HN points • 07 Feb 25
🕹 Technology Software Engineering
  1. You can create an API reference that adapts to different users, offering both a human-friendly and machine-readable version. It's important to meet the needs of both audiences.
  2. Using an OpenAPI document makes it easy to generate a comprehensive API reference without much effort. It's like having a complete guide available for your API.
  3. Content negotiation allows you to serve the right version of your API reference based on the request type. This way, humans get a readable document, while machines receive the necessary JSON data.

HTTP 204 Is the Best DELETE Response

The API Changelog • 0 implied HN points • 04 Jun 25
🕹 Technology Software Engineering
  1. HTTP 204 is a good response for DELETE operations because it means the action was successful and there's no further info needed. An empty response is often the best way to say everything worked out.
  2. Some people believe that a DELETE operation should include details about what was deleted, but that's not always necessary. You can get that info by checking before you delete.
  3. While 204 is recommended for DELETE actions, there are other options too. Situations may require different responses, but 204 often works best for clear communication.

Reading CSV files in UE5 C++

Load-bearing Tomato • 0 implied HN points • 23 Jul 25
🕹 Technology Software Engineering
  1. Using CSV files in UE5 can be tricky because the official documentation might not work as expected. It's important to double-check the methods for loading and parsing your data.
  2. To correctly read CSV files in UE5, use the 'LoadFileToString' method and then the 'CsvParser' module. This approach is confirmed to work, especially in version 5.3 and later.
  3. When writing CSV files, make sure to format your data properly with headers and ensure your output saves correctly. This process can save you frustration when managing your game data.

Edition 1: AI for Offense Is Here. Defenders Aren’t Ready.

Boring AppSec • 0 implied HN points • 17 Dec 25
🕹 Technology Software Engineering
  1. AI-orchestrated offensive campaigns are real and practical: coding agents, sub-agents and MCP can automate most of the cyber kill chain and run multi-day operations with minimal human input.
  2. Defenders are behind and must upskill quickly — learn to use AI defensively, run safe agent experiments in staging, assign dedicated AI-operator roles, and build human-in-the-loop checkpoints.
  3. AI tools bring new failure modes and risks: hallucinations mean you need verifier components, simple structured markdown can serve as a useful memory layer, and tight sandboxing plus MCP observability are critical to limit abuse.

An Interview With Jen Schreiber

ciamweekly • 0 implied HN points • 24 Nov 25
🕹 Technology Software Engineering
  1. CIAM should bridge the gap between security best practices and everyday users by making the secure choice the easiest default, using things like transparent MFA, just-in-time access, and session expiry to guide safe behavior.
  2. Modern CIAM is more complex and distributed across many systems and third parties, which widens the attack surface and makes rapid detection and response a core challenge.
  3. The future of CIAM is continuous, real-time access evaluation and automated response, with standards like the Shared Signals Framework enabling fast event sharing so access can be adjusted or revoked instantly.

AI and Drinking The Tech Kool-Aid

Synystron Synlogica • 0 implied HN points • 03 Dec 25
🕹 Technology Software Engineering
  1. Many tech trends and tools are fads that sweep the industry and later disappear. Don't automatically adopt something just because it's hyped.
  2. Relying on simple, proven tools and timeless techniques builds real, lasting skill. That practical focus helps you keep shipping useful work instead of chasing the latest craze.
  3. Experience and fundamentals give you a big-picture perspective that outlasts hype. Mastering core tools like the command line and editors such as vim is more valuable than following buzzword-driven practices.

Scope discipline when AI makes building fast

Bit Byte Bit • 0 implied HN points • 23 Dec 25
🕹 Technology Software Engineering
  1. Choose the right tool: build core, domain-specific messaging yourself and use SaaS like PostHog only where it clearly adds value (surveys, A/B tests).
  2. AI makes building fast and encourages scope creep, so keep your MVP narrow, put extras on an ideas list, and only implement features that solve the current problem.
  3. Don’t keep perfectly clean code you don’t need because it creates a maintenance burden. Use simple, flexible patterns (global LiveView hooks and small function-based rules) so you can extend behavior later without heavy rewrites.

Embedding-Based Tool Selection for AI Agents

Bit Byte Bit • 0 implied HN points • 21 Dec 25
🕹 Technology Software Engineering
  1. Embed tool descriptions and use semantic search to pick the top few relevant tools per query so you dramatically cut token usage and improve the model's tool‑selection accuracy.
  2. Choose an embedding provider based on your needs — calling OpenAI is simple and cheap for small volumes, while running a local model gives privacy and low latency but adds operational overhead — and hide that choice behind a provider abstraction so you can swap easily.
  3. Pure similarity can miss multi‑step dependencies, so expand selections by category and tune your similarity threshold, have a cold‑start fallback, and you'll get big wins in cost and latency.

@adlrocha - Taming the Agents: My "Spec-Test-Lint" Workflow for AI Coding

@adlrocha Weekly Newsletter • 0 implied HN points • 25 Jan 26
🕹 Technology Software Engineering
  1. Use a CLI-first setup with a terminal emulator and multiplexer, dedicating panes for the coding agent, an agent inbox, and the code to keep workflows fast and focused.
  2. Follow a Spec-Test-Lint cycle: start from a clear, exhaustive spec, set up tests and strict linting/CI up front, and write tests before or alongside code so each feature is fully tested and production-ready.
  3. Apply the same workflow to both quick side projects and complex codebases by varying strictness, keep a human-in-the-loop, sandbox experiments, and use agent-steering practices to reduce context switching and maintain quality.

🐿️ Danger of Death - The Insanely Profitable Tech Newsletter

Squirrel Squadron Substack • 0 implied HN points • 09 Feb 26
🕹 Technology Software Engineering
  1. When software can cause physical harm, use multiple layers of automated and human checks and avoid risky release practices.
  2. Many teams apply safety-critical processes to low-risk products and end up polishing for months, which wastes time and yields diminishing returns.
  3. Focus your engineers on finding and building what users actually need and will pay for, rather than protecting against unlikely catastrophic scenarios.

Using AI to Improve AI Tools

Front Left • 0 implied HN points • 17 Feb 26
🕹 Technology Software Engineering
  1. Use AI to build AI tools so those tools can iteratively improve themselves, removing the human as the weakest link in keeping systems up to date.
  2. Having tools that can self-audit and regenerate parts like knowledge synthesis and skill-writing creates a strong dogfooding loop that drives steady improvement.
  3. Be careful: large language models are stochastic, so recursive self-improvement won’t always converge and can spiral; set stopping rules and watch for diminishing returns.

Write Your APIs Like You Write a Movie

On Engineering • 0 implied HN points • 03 Mar 26
🕹 Technology Software Engineering
  1. Design your API as a clear workflow for AI actors by exposing chunky, outcome-focused tools instead of only low-level endpoints the agent must orchestrate.
  2. Make schemas, names, parameter descriptions, and especially error responses self-contained and consistent so an agent knows what happened, why it happened, and exactly what to do next.
  3. Test with real agents and multiple models, measure hallucinations and wrong-order calls, and be willing to redesign APIs for agent consumption rather than just wrapping existing endpoints.

Your Boss Said "Use AI." That's It. That's the Strategy.

On Engineering • 0 implied HN points • 22 Feb 26
🕹 Technology Software Engineering
  1. Top-down 'use AI' mandates fail without tooling, standards, or metrics. Ask for clarity and record a baseline metric now so you control the narrative.
  2. The engineer role is shifting from writing everything to orchestrating and validating AI output. Automate boring drudgery with AI but keep the parts of the job that require judgment and craft.
  3. Do practical, team-led experiments: run week-long spikes, build and share a prompt library, and practice in small dojos to learn when to trust or override AI. Measure PR review time and prioritize decision quality over raw speed.

The Tacit Ceiling of AI Knowledge Synthesis

Front Left • 0 implied HN points • 02 Mar 26
🕹 Technology Software Engineering
  1. Document synthesis hits a tacit ceiling because written sources mainly capture explicit knowledge, not the judgment and intuition experts use, so skills built from them often fail on edge cases and novel situations.
  2. Extraordinary quality requires extracting structural rationale and conceptual models — decision principles like “When X, do Y, because Z” — and using a Decision Skeleton that links triggers, choices, failure modes, and boundaries to turn knowledge into reliable actions.
  3. Pipeline safeguards (compression guards, critical-distinctions registries, adversarial tests, and iterative passes) improve results but can’t fully solve selection or recover tacit knowledge, so external domain expertise and objective validation remain necessary.

There are two systems. Which one are you working on?

Bad Software Advice • 0 implied HN points • 25 Mar 26
🕹 Technology Software Engineering
  1. You work on more than just the technical code — the system includes users, support, competitors, and the market, and missing that context can make your work irrelevant, wrongly specified, or badly prioritized.
  2. AI is lowering the cost of development, so developers are shifting from hand-coding everything to managing tools and judging agent outputs, which requires higher-level skills beyond writing code.
  3. Spend time learning the greater system and move up the stack; understanding users, support, and market forces helps you build the right thing and make better tradeoffs.