architecture

25 Aug 2026

When retrieval lies politely — working notes

Continue Reading

24 Aug 2026

Multi-agent theater vs useful specialization under real load

Continue Reading

07 Aug 2026

Connectors are the hard part of retrieval under real load

Continue Reading

22 Jul 2026

Owning outcomes when agents ship changes

Continue Reading

06 Jul 2026

RAG that fails gracefully when context is thin

Continue Reading

25 Jun 2026

Connectors are the hard part of retrieval

Continue Reading

20 Jun 2026

Multi-agent theater vs useful specialization

Continue Reading

19 Jun 2026

When retrieval lies politely (2026)

Continue Reading

12 Jun 2026

Grounding answers in systems teams actually run

Continue Reading

30 May 2026

Orchestration without a black-box conductor: a field note

Continue Reading

17 May 2026

Permissions in RAG: the boring requirement

Continue Reading

15 May 2026

Idempotency for agentic side effects for engineering leaders

Continue Reading

09 May 2026

Eval loops for retrieval quality

Continue Reading

27 Apr 2026

When an agent should stop and ask

Continue Reading

20 Apr 2026

When retrieval lies politely

Continue Reading

10 Apr 2026

Tool use as a privilege, not a default

Continue Reading

06 Mar 2026

RAG pipelines as team infrastructure without the theater

Continue Reading

03 Mar 2026

Idempotency for agentic side effects under real load

Continue Reading

31 Jan 2026

Event-driven agentic workflows teams can operate

Continue Reading

28 Jan 2026

From demo RAG to operable RAG for engineering leaders

Continue Reading

19 Jan 2026

Designing hand-offs between agents and humans — note 180

Continue Reading

12 Jan 2026

RAG pipelines as team infrastructure (2026)

Continue Reading

19 Dec 2025

Designing hand-offs between agents and humans when the calendar is full

Continue Reading

12 Dec 2025

RAG pipelines as team infrastructure — note 167

Continue Reading

30 Nov 2025

Agent harnesses: logs, budgets, and kill switches under real load

Continue Reading

20 Nov 2025

Orchestration without a black-box conductor

Continue Reading

21 Oct 2025

Idempotency for agentic side effects

Continue Reading

16 Oct 2025

Designing hand-offs between agents and humans without the theater

Continue Reading

01 Oct 2025

RAG pipelines as team infrastructure in practice

Continue Reading

28 Sep 2025

Who owns the knowledge graph behind the bot? — working notes

Continue Reading

03 Sep 2025

From demo RAG to operable RAG (2025)

Continue Reading

25 Aug 2025

From demo RAG to operable RAG

Continue Reading

24 Aug 2025

Who owns the knowledge graph behind the bot?

Continue Reading

22 Aug 2025

Designing hand-offs between agents and humans

Continue Reading

06 Aug 2025

Agent harnesses: logs, budgets, and kill switches

Continue Reading

26 Jul 2025

RAG pipelines as team infrastructure

Continue Reading

15 Jun 2025

Architecture decisions you can reverse (2026)

Continue Reading

13 Jun 2025

Architecture decisions you can reverse

Continue Reading

03 May 2025

Documentation that earns its keep — note 88

Continue Reading

28 Apr 2025

Code review as teaching, not gatekeeping — note 86

Continue Reading

26 Apr 2025

Testing strategy when speed matters (2024)

Continue Reading

25 Apr 2025

API design for humans who inherit it: a field note

Continue Reading

24 Mar 2025

Incident response that improves the system

Continue Reading

22 Feb 2025

Small platforms that stay operable

Continue Reading

18 Feb 2025

Technical debt with a repayment schedule

Continue Reading

11 Feb 2025

Migrations that respect team capacity (2025)

Continue Reading

22 Jan 2025

Testing strategy when speed matters

Continue Reading

18 Jan 2025

Documentation that earns its keep

Continue Reading

07 Jan 2025

Code review as teaching, not gatekeeping under real load

Continue Reading

12 Nov 2024

Migrations that respect team capacity

Continue Reading

16 Sep 2024

API design for humans who inherit it

Continue Reading

01 Sep 2024

Code review as teaching, not gatekeeping

Continue Reading