BestHub
Sign in
Tagged articles

LLM Engineering

18 articles · Page 1 of 1
Tencent Technical Engineering
Tencent Technical Engineering
Sep 21, 2026 · Artificial Intelligence

Jev Engineering: Moving Judgment Tasks Out of LLMs for Faster AI Agents

This article explores Jev (System One Model) as a specialized probability judgment layer for AI agents, detailing its three primitives (Choice, Score, Noul), the fast-jev-compaction project for context pruning, and a demo implementation showing 91.5% context reduction while preserving critical debugging details like file paths and error codes.

AI agentsJevLLM Engineering
0 likes · 32 min read
Jev Engineering: Moving Judgment Tasks Out of LLMs for Faster AI Agents
Tencent Cloud Developer
Tencent Cloud Developer
Sep 17, 2026 · Artificial Intelligence

Slashing Token Costs in Multi-Agent Workflows: Context Lifecycle & Batch Editing

Tencent's DevFlow multi-agent system cut token usage by 25–42% on a real 6-interface task by shortening context lifecycles with ephemeral code-exploration agents, progressive template loading, and a transactional replace_batch tool that merges multiple file edits into a single model round-trip.

DevFlowLLM Engineeringagent orchestration
0 likes · 26 min read
Slashing Token Costs in Multi-Agent Workflows: Context Lifecycle & Batch Editing
PaperAgent
PaperAgent
Sep 6, 2026 · Artificial Intelligence

Anthropic's Killer Multi-Agent Blueprint: One Loop, Skills, Harness & Snapshot Eval

Anthropic's production e-commerce and math-formalization agents share a unified architecture: a single-model loop with modular skills, tool calls to existing systems, code-enforced harness rules, and snapshot-based evaluation, enabling scalable, verifiable multi-agent systems.

Agent ArchitectureAnthropicLLM Engineering
0 likes · 17 min read
Anthropic's Killer Multi-Agent Blueprint: One Loop, Skills, Harness & Snapshot Eval
TechVision Expert Circle
TechVision Expert Circle
Jul 17, 2026 · Artificial Intelligence

Building Trustworthy AI Systems: Core Dimensions and Practical Solutions

The article outlines a comprehensive engineering approach for trustworthy AI, detailing five measurable dimensions—safety, reliability, explainability, privacy, and fairness—along with architecture design, input/output safeguards, hallucination mitigation, monitoring metrics, human‑in‑the‑loop strategies, and real‑world trade‑off recommendations.

AI safetyLLM EngineeringRAG
0 likes · 13 min read
Building Trustworthy AI Systems: Core Dimensions and Practical Solutions
AI Engineering
AI Engineering
Jun 22, 2026 · Artificial Intelligence

How Sakana’s Unconventional AI Orchestrator Fugu Beats Fable 5 in Code Benchmarks

Japanese startup Sakana’s new multi‑agent orchestration system, Fugu, combines publicly available models to deliver code‑generation performance that surpasses closed‑source rivals like Fable 5, offering two versions, detailed benchmark results, qualitative use‑case demos, pricing options, and an analysis of its engineering trade‑offs.

AI orchestrationFuguLLM Engineering
0 likes · 9 min read
How Sakana’s Unconventional AI Orchestrator Fugu Beats Fable 5 in Code Benchmarks
DataFunTalk
DataFunTalk
Jun 8, 2026 · Artificial Intelligence

How Harness Transforms AI Coding for Data Warehousing into an End-to-End Pipeline

This article details how a data‑warehouse team built a seven‑layer Harness framework to overcome AI‑coding challenges—semantic drift, strict constraints, and cross‑session context—enabling reliable, end‑to‑end production‑grade wide‑table delivery with up to 25× speedup and near‑zero side‑effects.

AILLM Engineeringautomation
0 likes · 32 min read
How Harness Transforms AI Coding for Data Warehousing into an End-to-End Pipeline
Java Tech Enthusiast
Java Tech Enthusiast
May 26, 2026 · Artificial Intelligence

Why Interviewers Should Ask About Harness Engineering – Distinguishing It from Prompt and Context Engineering

The article explains how AI is evolving from simple chat interactions to production‑grade workflows by progressing through Prompt Engineering, Context Engineering, and finally Harness Engineering, detailing their distinct goals, practical examples, step‑by‑step processes, and why Harness is essential for building controllable, auditable AI systems.

AI workflowContext EngineeringHarness engineering
0 likes · 21 min read
Why Interviewers Should Ask About Harness Engineering – Distinguishing It from Prompt and Context Engineering
Machine Learning Algorithms & Natural Language Processing
Machine Learning Algorithms & Natural Language Processing
Apr 28, 2026 · Artificial Intelligence

Why DeepSeek V4 Insists on Batch Invariance—and What It Costs

DeepSeek V4 achieves ultra‑long context, complex training pipelines, and custom high‑performance kernels by enforcing batch invariance, a design that guarantees bit‑wise identical outputs across varying batch shapes but incurs lower GPU utilization, reduced small‑batch speed, and added engineering complexity.

Batch InvarianceDeepSeek-V4GPU utilization
0 likes · 8 min read
Why DeepSeek V4 Insists on Batch Invariance—and What It Costs
AI Architecture Hub
AI Architecture Hub
Apr 21, 2026 · Artificial Intelligence

Why Harness Architecture Turns LLMs into Production‑Ready Agents

This article explains why the Harness architecture—linking prompts, context, and runtime support—is the decisive factor that turns large language models from demo prototypes into reliable production agents, detailing its core capabilities, structural components, execution loop, design trade‑offs, and industry trends.

AI operationsLLM EngineeringProduction AI
0 likes · 35 min read
Why Harness Architecture Turns LLMs into Production‑Ready Agents
Code Archaeology
Code Archaeology
Mar 26, 2026 · Artificial Intelligence

How Deep Agents Implements Memory: AGENTS.md Injection as Procedural Memory

This article maps cognitive psychology's semantic, episodic, and procedural memory to AI agent architecture, then details Deep Agents' MemoryMiddleware implementation that loads AGENTS.md files at initialization and injects their content into system prompts, functioning as procedural memory with semantic elements.

AGENTS.mdAI agentsDeep Agents
0 likes · 10 min read
How Deep Agents Implements Memory: AGENTS.md Injection as Procedural Memory
Code Archaeology
Code Archaeology
Mar 23, 2026 · Artificial Intelligence

How Deep Agents Implements Skills: Progressive Disclosure & Middleware Ordering

This article analyzes Deep Agents' Skills implementation, revealing a three-layer progressive disclosure loading pattern where metadata loads at startup, full instructions load on activation, and resources load on demand. It explains how before_agent and wrap_model_call hooks inject skill metadata into system prompts, and why SkillsMiddleware positioning differs between main agents (early for coordination) and sub-agents (late for separation of concerns).

AI agentsAgent ArchitectureDeep Agents
0 likes · 12 min read
How Deep Agents Implements Skills: Progressive Disclosure & Middleware Ordering
Code Archaeology
Code Archaeology
Mar 18, 2026 · Artificial Intelligence

Deep Agents Context Management: How SummarizationMiddleware Auto-Compresses History

This article details Deep Agents' context management implementation using SummarizationMiddleware, which automatically compresses conversation history during model calls by offloading old messages to backend storage and replacing them with generated summaries, including threshold configuration, cutoff determination, and the three message views that enable traceable, iterative summarization.

Agent MiddlewareBackend StorageConversation Summarization
0 likes · 11 min read
Deep Agents Context Management: How SummarizationMiddleware Auto-Compresses History
AI Architecture Hub
AI Architecture Hub
Mar 15, 2026 · Artificial Intelligence

How OpenClaw Solves Long‑Task Context Challenges for AI Agents

This article analyses the real‑world pain points of long‑running AI agents, breaks down OpenClaw’s core concepts, explains its three‑layer context‑compression pipeline, presents four key engineering decisions, shares six practical techniques with essential parameters, and compares OpenClaw to competing approaches.

AI agentsAgent ReliabilityLLM Engineering
0 likes · 17 min read
How OpenClaw Solves Long‑Task Context Challenges for AI Agents
Code Mala Tang
Code Mala Tang
Mar 9, 2026 · Artificial Intelligence

How Claude’s New Prompt Caching Cuts Token Costs by 90% for Long‑Running Agents

Claude’s API now automatically caches static parts of prompts—system instructions, tool definitions, and context—so repeated calls reuse these sections at only 10% of the standard token price, dramatically reducing costs for multi‑turn agents, but developers must manage prefixes and avoid cache‑breaking changes.

Claude APILLM EngineeringPrompt Caching
0 likes · 15 min read
How Claude’s New Prompt Caching Cuts Token Costs by 90% for Long‑Running Agents
Code Archaeology
Code Archaeology
Jan 26, 2026 · Artificial Intelligence

OpenAI vs Anthropic Tool Calling APIs: Conversation vs Execution Trajectory Abstractions Compared

This article compares OpenAI's Chat Completions and Responses APIs with Anthropic's Messages API for tool calling, detailing request/response structures, abstraction differences between conversation and execution trajectory models, and Anthropic's advanced features including tool search, programmatic tool calling, and tool use examples for large-scale agent systems.

API designAgent EngineeringAnthropic
0 likes · 19 min read
OpenAI vs Anthropic Tool Calling APIs: Conversation vs Execution Trajectory Abstractions Compared
dbaplus Community
dbaplus Community
Jan 21, 2026 · Information Security

How Large Language Models Transform Data Security: Frameworks, Challenges, and Real-World Practices

This article reviews the current state, feasibility, industry adoption, concrete deployment scenarios, and future directions of applying large language models to data security, covering technical challenges, architectural designs, prompt engineering, privacy‑preserving techniques, and practical case studies.

AI applicationsData SecurityLLM Engineering
0 likes · 21 min read
How Large Language Models Transform Data Security: Frameworks, Challenges, and Real-World Practices
Code Archaeology
Code Archaeology
Dec 22, 2025 · Artificial Intelligence

RAG Explained: Core Concepts, Architecture, and a Minimal LangChain Demo

This article introduces Retrieval-Augmented Generation (RAG), covering its origins, core components (retriever and generator), advantages like reduced hallucination and knowledge updatability, architectural challenges such as chunk sizing and vector search algorithms, and provides a step-by-step LangChain implementation using FAISS, SentenceTransformer, and DeepSeek.

FAISSKnowledge RetrievalLLM Engineering
0 likes · 15 min read
RAG Explained: Core Concepts, Architecture, and a Minimal LangChain Demo