Standard Reader

AI Engineering

Thoughts on how to thrive in the age of vibe coding, using AI agents, how to engineer with and for AI.

@rajesh-vasa.bsky.social8readers26posts3mo ago

LatestRecent writing

The Agent, Not the Task

Compression, Adaptation, and the Limits of Scalar Reward

The logic of saying "no"

What Anthropic's Pentagon standoff reveals about trust in AI is that refusal can be a strategy in the age of AI.

Beyond the encoding gap: Metafeatures, Swarms, and the Architecture of Discovery

On artificial discovery and its human costs

Beyond the encoding gap: Metafeatures, Swarms, and the Architecture of Discovery

On Artificial Discovery and Its Human Cost

The Residue and the Trace: What AI Cannot Extract from Expert Knowledge

Tacit Knowledge, Training Corpus Boundaries, and the Knowledge AI Cannot Reach

Why AI can make organisations slower before it makes them faster

AI dramatically accelerates cognitive production while leaving coordination, judgment, and context-building at human speed. The result is not acceleration but fragmentation. Bottlenecks migrate to integration points we have not designed for. The path forward requires building epistemic scaffolding, deliberate backpressure, and systems that treat human judgment as the scarce resource it has/will become.

The Archaeological Reversal

When AI sees structure better than you, what can we humans do?

The Composition Problem

The spec-verify loop works for parts, but the theory of how parts relate lives in no single specification. Composition requires its own genesis, its own traversal, at the level of interaction between stable components.

The Encoding Gap

Why working code is not the same as understanding. What is the human role in the age of coding agents? and Why it is not as straight-forward as 'just use coding agents' and it will all be fine.

Augmenting Human Intellect in the Age of Agentic AI: Revisiting Engelbart Concepts from 1960s

What Changes When Our Tools Become Agents

Creating reliable Systems from unreliable parts: Thirty Years of Designing Around Uncertainty

Why perfecting AI components is the wrong goal; Orchestration over perfection.

The world is full of isomorphisms

Starting to feel a sense of deja-vu scrolling through new all the news about AI? Does it all seem to be the same stuff over and over about AI across LinkedIn, X, Discord, Reddit?