Reference notes and collected resources on Videos Tutorials.

Videos Tutorials

Engineering articles, software architecture research, and personal notes by Adam Manuel.

Adam Manuel’s “Dev Blog”

An Express.js architecture tutorial covering domain-driven structure, dependency injection, databases, middleware, services, and testing.

My Favorite Express Architecture

Mastra is an open-source TypeScript framework designed to help developers build AI applications and features quickly.

Guide to Building Autonomous AI Agents

In today’s fast-paced academic landscape, staying current with the latest research can be a daunting task.

Let ChatGPT Be Your Research Asst.

The sources and our conversation history offer several solutions for tackling the challenges of data sparsity and cold start when employing deep learning for recommendation tasks:

Addressing Data Sparsity and Cold Start in Deep Learning Recommenders

Academic labs, papers, and research resources for recommendation engines.

Recommendation Engines Academic  Paper Sources

Building a reliable software platform requires more than just writing code—it demands a structured approach to testing and a well-thought-out PR workflow.

Refining Testing Strategies and PR Workflows for Scalable Development

In the fast-paced world of job hunting, staying ahead of the competition means being the first to know about new opportunities.

Speed Up Your Job Hunt With This LinkedIn Scraper and Your System Notifications

Practical CommonJS-to-TypeScript migration notes on utility types, type-only imports, JSDoc, and constructor dependency injection.

CommonJS →  Typescript Migration Tips

Research resources on project management, programming languages, and software engineering practices.

Project Manag. Research Topics

Reference notes and collected resources on Laws of UX.

Laws of UX

Generate React components from SVG URLs or local files using SVGR and open the result in VS Code from the terminal.

Creating SVG React Component with a single terminal command

Explore package.json from the terminal with the fx JSON explorer.

package.json “—help” CLI

MongoDB research notes on consistency, schema design, indexes, shard keys, and write performance.

MongoDB Optimization

A twelve-factor app is always tracked in a version control system, such as Git, Mercurial, or Subversion.

12 Factor Application

Microservices are an extension of service oriented architectures.

Domain-Oriented Microservice Architecture 

I talked about the importance of reading foundational papers last week.

Foundational Distributed Systems Papers

Further distributed-systems papers on consensus, replication, scalable data stores, and fault tolerance.

Additional Distributed Systems Papers

This README containts OpenIE and ORE papers and resources.

Open Information Extraction and Open Relation Extraction Papers

Reference notes and collected resources on Knowledge Graph Tools.

Knowledge Graph Tools

The funding and burn rate progression spans approximately 6 years, with each funding round leading to increased investment and a targeted percentage goal.

Optimal Fund Raising & Burning Rates

Reference notes and collected resources on Optimizing Performance.

Optimizing Performance

Reference notes and collected resources on Building Engineering Organizations.

Building Engineering Organizations

Reference notes and collected resources on Big Data.

Big Data

Reference notes and collected resources on Distributed Machine Learning.

Distributed Machine Learning

Reference notes and collected resources on Distributed Locking.

Distributed Locking

Reference notes and collected resources on Stabilizing Systems.

Stabilizing Systems

An AI Agent is an intelligent software system that can autonomously perform tasks, make decisions, and take actions on behalf of users or organizations.

Help Center

A step-by-step tutorial for setting up a scalable, enterprise-ready Express.js application architecture from scratch.

My Favorite Express Architecture Setup Guide

Official MongoDB resources on performance, schema patterns, transactions, courses, tutorials, and data pipelines.

MongoDB Official Documentation

Below is a brief description of each of the schema design anti-patterns we've covered in this series.

Schema Design

For compound indexes, this rule of thumb is helpful in deciding the order of fields in the index:

MongoDB Indexes

Research notes on choosing MongoDB shard keys, including compound hashed keys, distribution, and targeted queries.

Choosing a shard key for MongoDB

Optimizing write performance helps you get the most out of Azure Cosmos DB API for MongoDB's unlimited scale.

Optimize for write performance with Cosmos DB using the MongoDB API

Operations in a transaction use the transaction-level read preference.

Read Concern - Write Concern - Read Preference - What does it mean?

Defining microservice boundaries is probably the first challenge anyone encounters.

Challenges and solutions for distributed data management

The most powerful part about aggregates is that they dispatch Domain Events which can be used to co-locate business logic in the appropriate subdomain.

Domain Driven Design Models

The goal when identifying model boundaries and size for each microservice isn't to get to the most granular separation possible, although you should tend toward small…

Identifying domain-model boundaries for each microservice

Martin Kleppmann’s paper on making conflict-free replicated data types resilient to Byzantine faults.

Making CRDTs Byzantine Fault Tolerant

The Sage Agent is an advanced AI assistant designed to provide thoughtful, comprehensive responses across a wide range of topics.

More Details Sage Agent

The ATLAS Agent is an advanced AI-powered system designed to streamline complex workflows and enhance productivity across various domains.

More Details ATLAS Agent

Aria Agent is an advanced AI-powered assistant designed to streamline workflows and enhance productivity.

More Details Aria Agent

Zephyr Agent is an advanced AI-powered automation tool designed to streamline workflows and enhance productivity.

More Details Zephyr Agent

The Pace Agent is an intelligent assistant designed to help users manage their workflow and productivity by monitoring task completion rates and providing adaptive recommendations.

More Details Pace Agent

The Kulshan Agent is an AI-powered assistant designed to streamline workflows and enhance productivity.

More Details Kulshan Agent

MongoDB performance resources covering data modeling, memory sizing, indexes, and deployment practices.

Best Practices

MongoDB Atlas and MongoDB University courses on deployment performance and optimization.

Full Courses from Atlas

MongoDB schema design patterns and the data-modeling problems they address.

Schema Patterns

Background reading on MongoDB transactions, WiredTiger timestamps, and multi-document ACID guarantees.

Transactions Background

Resources on integrating MongoDB with Kafka and building data pipelines.

Big Data Pipeline

Queries that contain the shard key and can be sent to a single shard or subset of shards are called targeted queries.

Sharding Details

Why splitting a slow task did not remove the bottleneck.

I Split the Background Jobs. They Still Waited on Each Other.

Why preserving behavior includes the interface people already know.

The Refactor Passed Its Checks. The Product Still Changed.

Why an AI behaving correctly during a test can still leave the main question unanswered.

If You Cannot Reproduce the Bug, You Cannot Prove the Fix

Make customer isolation a rule enforced when data is read or changed, then test the boundaries directly.

Hiding a Button Does Not Protect the Data

A reassuring dashboard number can answer the wrong question.

What Does Your Success Rate Actually Count?

Separate product decisions from rendering so a frontend migration does not rewrite the business.

Business Rules Should Survive a UI Rewrite

Use a clear promise, explicit exclusions, and a working slice to control scope under a deadline.

Build a Complete Small Product Before an Incomplete Big One

Preserve missing evidence instead of turning it into a confident negative answer.

Unknown Is Not False: Designing Honest AI Results

Design and test restoration around active writers, related files, crash points, and evidence that the recovered product works.

A Database Backup Is Not Yet a Recovery Plan

How a scheduling rule can produce a surprising result.

Adding More Availability Can Make a Time Slot Disappear

Why passing tests still need an independent definition of correct behavior.

The Same AI Can Write the Code, the Tests, and the Same Mistake Three Times

Give scheduled notifications a stable identity so delayed results and retries do not create extra messages.

An Hourly Digest Needs More Than an Hourly Timer

A download investigation that required checking beyond the status label.

The File Said Uploaded. The File Was Missing.

Why cached AI results need to remember the inputs and rules that produced them.

Your Cache May Be Answering Yesterday's Question

Why long debugging efforts need a fresh view of what is failing now.

Stop Fixing Yesterday's Incident

Turn recurring engineering problems into small, testable improvements without confusing activity with impact.

Measure the Friction You Want to Remove

Design product states that distinguish user choices, external facts, and stale interface actions.

A Status Label Is a Claim About What Happened

Split work around stable boundaries and shared decisions so extra contributors reduce elapsed time.

Parallel Engineering Depends on Clear Ownership

Measure what costs money, change one thing, and check that the product still works.

How I Would Cut a Cloud Bill Without Guessing

Explain what a polished demonstration proves, what it simulates, and what still needs validation.

A Great Demo Still Needs an Evidence Label

A simple way to organize a growing app without creating a junk drawer.

Refactoring React Around Features: Where Should Shared Code Live?

Why simplifying a queue system requires understanding ownership and recovery.

When Two Queues Disagree About the Same Job

A small experiment for learning whether an AI feature is useful.

Testing an LLM Integration on a $10 Budget

How watching a user flow can reveal questions your automated checks never asked.

What a Screen Recording Reveals That a Passing E2E Test Doesn't

What happens when two services disagree about the meaning of a number.

How Can a Job Be Billions of Percent Complete?

A practical test for whether project knowledge is actually transferable.

Could Another Developer Take Over Without a Meeting?

Why document accessibility requires checking reading order as well as appearance.

A Table Can Look Right and Read Wrong

A practical way to turn code and Git history into questions worth investigating.

What I Want an LLM to Tell Me About an Unfamiliar Codebase

Make automation predictable with clear inputs, structured output, and visible outcomes.

Designing a Command-Line Tool an AI Can Actually Use

A first-week plan for finding a small improvement that matters to the team.

Make yourself comfortable

Find your next rabbit hole

Videos Tutorials

Throughput Calculations