BestHub
Sign in
Architect Practice
Author

Architect Practice

Committed to sharing tech and documenting ideas.

35
Articles
0
Likes
135
Views
0
Comments
Recent Articles

Latest from Architect Practice

35 recent articles
Architect Practice
Architect Practice
Oct 13, 2026 · Artificial Intelligence

How AI Engineers Decide What to Build: Four Skills for Shaping the Build

Andrew Ng's final AI Engineering Skills Map installment defines four capabilities for engineers to shape what gets built: driving rapid build loops with a bias for action, making product decisions grounded in user empathy, communicating across functions to align stakeholders, and exercising high-agency ownership to judge what's worth building.

AI engineeringAndrew Ngbuild loop
0 likes · 14 min read
How AI Engineers Decide What to Build: Four Skills for Shaping the Build
Architect Practice
Architect Practice
Oct 6, 2026 · Artificial Intelligence

From Coder to Agent Orchestrator: Mastering AI Coding Agents

Andrew Ng outlines a three-step iterative workflow (Plan, Execute, Deploy & Monitor) and five core capabilities—directing workflows, enabling autonomy, verifying output, customizing environments, and understanding failure modes—for AI engineers to effectively steer coding agents, emphasizing that judgment outweighs tool operation as coding becomes cheaper.

AI engineeringAndrew NgCoding Agents
0 likes · 18 min read
From Coder to Agent Orchestrator: Mastering AI Coding Agents
Architect Practice
Architect Practice
Oct 5, 2026 · Artificial Intelligence

AI Writes Code, But Misses Trade-offs: Why Software Engineering Fundamentals Are Essential

Andrew Ng argues that software engineering fundamentals are more crucial than ever for AI engineers because coding agents can implement solutions but cannot recognize critical trade-offs in latency, reliability, security, and data architecture; engineers must understand full-stack systems, data management, architecture evolution, security shift-left, and production operations to guide AI effectively.

AI engineeringAndrew NgCoding Agents
0 likes · 15 min read
AI Writes Code, But Misses Trade-offs: Why Software Engineering Fundamentals Are Essential
Architect Practice
Architect Practice
Sep 25, 2026 · Industry Insights

Andrew Ng's AI Engineering Skills Map: 4 Core Skills from 10K+ Job Analyses

Andrew Ng analyzed 10,000+ job descriptions and expert interviews to identify four multiplicative core skills for AI engineering: building/deploying AI applications, software engineering fundamentals, using coding agents, and shaping build direction, emphasizing that missing any one skill nullifies overall effectiveness.

AI engineeringAndrew NgCoding Agents
0 likes · 15 min read
Andrew Ng's AI Engineering Skills Map: 4 Core Skills from 10K+ Job Analyses
Architect Practice
Architect Practice
Sep 7, 2026 · R&D Management

Passing Acceptance Isn't About More Documents—It's About Decision Ownership

This article explains why traditional software projects fail acceptance despite complete documentation, emphasizing that each valid document must have an owner, version, and decision traceability. It distinguishes state vs. decision documents, details critical document pairs, shows how changes must propagate through the entire document chain, and provides a checklist to align documentation with contractual and operational needs.

change managementconfiguration managementdocumentation management
0 likes · 30 min read
Passing Acceptance Isn't About More Documents—It's About Decision Ownership
Architect Practice
Architect Practice
Aug 20, 2026 · Artificial Intelligence

From a World Championship Win to ChatGPT: What OpenAI Got Right in Seven Years

The article traces OpenAI’s seven‑year journey from the OpenAI Five Dota 2 victory to ChatGPT, showing how a clear goal, massive self‑play, scaling of compute and data, and systematic transfer of learned capabilities enabled the transition from game‑playing AI to a widely used conversational product.

AI researchChatGPTDota 2
0 likes · 22 min read
From a World Championship Win to ChatGPT: What OpenAI Got Right in Seven Years
Architect Practice
Architect Practice
Aug 1, 2026 · Industry Insights

Why Enterprise Data Intelligence Must Evolve from Text‑to‑SQL to Data Agents

The article traces four generations of enterprise data intelligence—from BI dashboards to Text‑to‑SQL and finally Data Agents—explaining why successful deployment depends on five engineering disciplines (Context, Knowledge, Skill, Memory, Evaluation Loop) rather than the underlying LLM model.

Data AgentEnterprise Data IntelligenceEvaluation Loop
0 likes · 18 min read
Why Enterprise Data Intelligence Must Evolve from Text‑to‑SQL to Data Agents
Architect Practice
Architect Practice
Jul 29, 2026 · Artificial Intelligence

How a 1.5B Model Beats Cutting‑Edge Large Models on Math Exams

This article explains why and how knowledge distillation lets a 1.5 B parameter model surpass much larger LLMs on math benchmarks, detailing the underlying soft‑label transfer, temperature tuning, various distillation families, engineering pipelines, and the practical trade‑offs that bound its success.

AI engineeringLarge Language Modelsknowledge distillation
0 likes · 14 min read
How a 1.5B Model Beats Cutting‑Edge Large Models on Math Exams
Architect Practice
Architect Practice
Jul 6, 2026 · Artificial Intelligence

How Cursor Retrieves a Specific Function from a 50,000‑File Repo in Under a Second

The article dissects Cursor's dual‑index architecture—semantic vector search built on AST chunking and a custom‑trained embedding model, plus a local trigram regex index—explaining how Merkle trees, team index reuse, and file‑based context delivery enable sub‑second code retrieval in massive monorepos.

AgentMerkle treeTurbopuffer
0 likes · 18 min read
How Cursor Retrieves a Specific Function from a 50,000‑File Repo in Under a Second
Architect Practice
Architect Practice
Jul 4, 2026 · Artificial Intelligence

Why Can an Agent Remember You? Exploring Memory Types and Retrieval Routing

The article analyzes why agents often forget or hallucinate past information, defines three orthogonal memory dimensions, critiques common naive solutions, and presents a four‑module architecture—including extraction, representation, retrieval, and maintenance—plus cross‑cutting concerns, a reference design, failure patterns, and a workload‑driven selection matrix.

Agent MemoryKnowledge GraphLLM
0 likes · 24 min read
Why Can an Agent Remember You? Exploring Memory Types and Retrieval Routing