Skip to content
channel · books & ideas

Book Cube

Books, ideas and engineering culture — every week

A Telegram channel with original notes and reviews on books about technology, architecture, management, economics and personal growth. Long-form posts with practical takeaways, no fluff, no ads.

Open channel on TelegramRelated materials
Public archive

Book Cube posts

Bilingual versions of channel posts, with sources and a link to the original Telegram publication.

  1. #Management

    Organizational Redesign for a Division of 1,000 Engineers

    A solo episode announcement on dividing a large engineering organization into five units and handing authority over to CTOs.

  2. #PlatformEngineering

    3 AImigo S1E13: Infrastructure for Agentic Development

    An episode announcement on how CI, review, and access management change under workloads created by agents.

  3. #Books

    Corporate [Trans]formation: Strengthening a Business by Developing People

    A review of the Borisovs' book on HR processes, employee learning, and management maturity.

  4. #AI4SDLC

    State of AI4SDLC: How Software Development Changes as Code Becomes Easier to Write

    The DotNext keynote recording, summary, and slides on delivery, migrations, and knowledge that should outlive the code.

  5. #Engineering

    [3/3] Octoverse 2026: Agents, Models, and Development Workflows

    Part three of a look at GitHub's report, covering delegation to agents, model selection, review, and security.

  6. #Book

    Code of Leadership: How to Lead Engineers by Example When Agents Write the Code

    A working plan for a 20-chapter book on a leader's practice, delegation, and change in engineering teams.

  7. #Engineering

    [2/3] Octoverse 2026: Languages, Industries, and Open Source

    Part two of a look at GitHub's report, covering languages, industry classification, and open-source growth.

  8. #Engineering

    [1/3] Octoverse 2026: Methodology, Scale, and Geography

    Part one of a look at GitHub's report, covering methodology, activity at scale, and community geography.

  9. #AI4SDLC

    Beyond the Model: When a Harness Helps an Agent

    A breakdown of a comparison of agent harnesses and NanoHarness experiments on ProgramBench.

  10. #AI

    3 AImigo S1E12 Materials: October 2026 AI Digest

    The October digest materials cover agent security, model economics, robotics, and checking results.

  11. #AI4SDLC

    Laurie Voss: What's Left for Engineers When AI Writes the Code?

    A talk breakdown on the future of engineering work and the author's view on the limits of agent autonomy.

  12. #AI4SDLC

    On Randomness in Agentic Evals: Noise in Agent Evaluations

    A breakdown of repeated SWE-bench Verified runs and notes on telling a real improvement from noise.

  13. #AI4SDLC

    Research Insights #43 Materials: Agent Skills and Open Knowledge Format

    Episode materials on testing agent skills, shared knowledge bases, and data provenance.

  14. #AI

    3 AImigo Live: October News Digest

    An invitation to the live stream of the 3 AImigo October news digest.

  15. #AI4SDLC

    AI Dev Live 2026: Did AI Speed Up Your Development?

    A talk announcement on measuring AI's effect on development flow and accounting for the full cost of a result.

  16. #AI4SDLC

    Code of Leadership #86 Materials: AI Agents and Developer Productivity

    Episode materials from a conversation with Vladimir Dolgov about metrics, planning, and trust in AI adoption analytics.

  17. #SystemDesign

    System Design Interview: Architectural Thinking in Big Tech

    A book announcement with a seven-step system design interview framework and a map of all 35 chapters.

  18. #AI4SDLC

    3 AImigo S1E12: October 2026 AI Digest

    A live stream announcement covering Anthropic's IPO, model costs, agent security, and AI in engineering workflows.

  19. #Data

    Andy Pavlo: What AI Agents Change for Databases

    An interview breakdown covering agent workloads, permissions, memory, database tuning, and the convergence of OLTP and OLAP.

  20. #AI

    3 AImigo S1E11 Materials: The AI-Native State

    Episode materials on automating government procedures, agent permissions, and the right to appeal decisions.

  21. #AI4SDLC

    Harness at Giga Conf: Talk Materials

    The recording, slides, and long read for a talk about which parts of an agent harness to reassess when the model changes.

  22. #AI4SDLC

    Evaluating AGENTS.md: What Context Is Still Missing?

    A look at how context files affect agent costs on CTXBench and SWE-bench Lite, and why task preparation may matter.

  23. #AI4SDLC

    Max Kanat-Alexander: do we need quality in code that can be replaced?

    Code quality, replaceable implementations, and the knowledge that must survive component regeneration.

  24. #AI4SDLC

    Research Insights #42 materials: agent evaluations—how to tell whether the agent succeeded

    Episode materials on outcomes, graders, pass@k and pass^k, regression checks, and agent monitoring.

  25. #AI4SDLC

    How Meta built Forum with AI agents: the article and my margin notes

    Building Forum with AI agents, with my notes on the SDLC, specifications, architecture, and prototyping.

  26. #AI4SDLC

    Research Insights #41 materials: anatomy of an agent—seven harness responsibilities

    Episode materials on seven harness responsibilities, state, permissions, and testing architectural ideas.

  27. #Personal

    A morning with views like these

    It is lovely when the morning starts with views like these.

  28. #Books

    Agentic Engineering — Addy Osmani

    The first five chapters of Osmani’s book: specs as a thinking tool, experience in the agent’s environment, and the cost of coordination.

  29. #Management

    Research Insights #40 materials: Anthropic—what remains for people in a company of agents

    Episode materials on human attention, learning, and responsibility in a company where agents do more and more.

  30. #Personal

    Medium finally adds dark mode

    An email about Medium’s long-awaited dark mode, and a little irony about timing.

  31. #AI

    [2/2] Stanford CME295: GLM-5 and the role of the agent harness

    How GLM-5 combines training stages, restores skills, and changes the requirements for an agent harness.

  32. #Personal

    UK Visa Approval and Plans for London

    My UK visa, my wife’s, and our youngest son’s have been approved; our middle son needs additional documents. The plan is to be in London in the second half of November.

  33. #AI4SDLC

    Agents Write Code. How Do We Know Development Has Become More Effective? | Code of Leadership

    An announcement of Code of Leadership with Vladimir Dolgov on developer productivity, the impact of AI agents, the cost of verification, and DevOptify metrics.

  34. #AI4SDLC

    Agent Skills and Open Knowledge Format (OKF): Skills, Knowledge, and Trust

    An announcement of a Research Insights Made Simple livestream with Maxim Smirnov about Agent Skills, Open Knowledge Format, skill evaluation, and trust in knowledge.

  35. #AI

    [1/2] Stanford CME295: How LLMs Are Trained

    Part one of a breakdown of Stanford CME295’s third lecture: how pretraining, SFT, preference tuning, RL, and on-policy distillation provide different training signals.

  36. #AI

    3 AImigo S1E10 Materials: Agents with Payment Authority

    Materials for the tenth 3 AImigo episode on shopping with AI agents: preferences, permissions, order approval, merchant incentives, and responsibility for mistakes.

  37. #AI4SDLC

    Demystifying evals for AI agents: Evaluation Structure and Margin Notes

    The structure of evals in Anthropic’s article, with my notes on TDD, expectations, the DOM, independent trials, and trajectory evaluation.

  38. #Leadership

    Code of Leadership #85 Materials: FDE with Pavel Golubev

    Pavel Golubev on delivering AI for customers: feasibility checks, business commitments, access permissions, and FDE teamwork.

  39. #AI4SDLC

    Agent evals: how to tell whether an agent succeeded / Research Insights Made Simple #42

    Research Insights Made Simple No. 42: checking agent outcomes, choosing graders, and distinguishing a better system from a higher score.

  40. #Humor

    The How to be British Collection Two

    Martyn Ford and Peter Legon’s second collection: British humor about food, railways, compliments, and self-deprecation.

  41. #Personal

    Research Insights #41 live: the anatomy of an agent

    Research Insights Made Simple No. 41 is starting, covering the seven responsibilities of a harness and the Harness Engineering preprint.

  42. #AI4SDLC

    Jediah Katz on the harness: what remains after a new model

    Jediah Katz’s talk on the co-evolution of models and harnesses: tracking temporary shims, testing their removal, and preserving authorization boundaries.

  43. #AI4SDLC

    Research Insights #39 materials: Scion and development with agents

    Materials for Research Insights Made Simple No. 39, October 15, 2026: Scion, server-side source control, the engineers’ role, and the limits of validation.

  44. #AI4SDLC

    Harness Engineering: the anatomy of coding agents

    A source-code study of 11 coding agents: seven harness subsystems, engineering trade-offs, and questions about the authors’ conclusions.

  45. #Personal

    Live: how AI is changing work at Anthropic

    An invitation to a live discussion of an Anthropic article on how using AI changes work inside the company.

  46. #DevEx

    Anthropic: What Remains for Humans in a Company of Agents? Research Insights Made Simple #40

    An announcement of a discussion of Anthropic’s research: outcomes, human attention, learning, and incentives.

  47. #AI4SDLC

    Research Insights №38: What a Harness Contributes — Materials

    Episode materials on experimentally evaluating agent planning, tools, and context management.

  48. #AI

    How Work Is Changing Inside Anthropic

    Research on hiring, productivity, meta-work, and employee incentives, with the author’s PDF annotations.

  49. #AI4SDLC

    Gautam Korlam: CI for a Stream of Agent Changes

    CI capacity, repeated checks, independent review, and retaining the evidence behind merge decisions.

  50. #SelfDevelopment

    Understand Before Deciding: How I Learn New Topics with AI

    A pmclub upskill week talk announcement about using AI to work with sources and decision criteria.

Continue in the full archive
What's inside · manual May 2026 snapshot
4,500+
posts
30+
topic tags
5 years
of regular posting
1
author · no editorial

Why subscribe

  • 01Get notes on key books with practical takeaways and links to the originals
  • 02See how the same idea echoes across engineering, management and economics
  • 03Be the first to hear about new talks, articles and podcast episodes
  • 04Keep a ready-to-use list of recommendations: 'what should I read on X'

How the channel works

Deep reviews

Reviews of books, research, and talks: key ideas, context, practical conclusions, and a personal perspective.

Topic tags

Posts are tagged: #Engineering, #Management, #AI, #Economics, #Software, #Leadership and more — easy to find everything on a topic.

Announcements & threads

New talks, tellmeabout.tech articles and Code of Leadership episodes are announced here too. One feed for everything.

Channel sections

Real tags from recent months — frequency reflects the channel's focus: tech, AI and management at the center, but not only.

01#Engineering
02#AI
03#Management
04#Software
05#Processes
06#Agents
07#Leadership
08#SystemDesign
09#Productivity
10#Architecture
11#Whitepaper
12#DistributedSystems
13#Database
14#Future
15#Economics
16#PopScience
17#Strategy
18#Math

Post genres

What you'll actually see in the feed.

01book review

Engineering book review

From 'Designing Data-Intensive Applications' to fresh AI engineering whitepapers — with a take on what's useful and for whom.

02talk announcement

Talk announcement

When a new conference talk comes out — recording, slides, references and a short recap land here.

03engineering note

Practice note

Short observations and ideas: how we build architecture, how culture changes, what works in large orgs.

04research

Whitepapers & research

Reviews of DORA, DevEx, AI4SDLC and other studies — with focus on what's actually applicable.

Who it's for

  • Engineers who want depth, not summaries of summaries
  • Team leads and managers looking for books in their growth zone
  • CTOs and founders holding both the tech and the business context
  • Anyone who prefers long, thoughtful posts over a news feed

Subscribe

One post every few days, zero ads, thousands of readers as of May 2026 — join if you learn through books.

Open Book Cube