BestHub
Sign in
DataFunTalk
Author

DataFunTalk

Dedicated to sharing and discussing big data and AI technology applications, aiming to empower a million data scientists. Regularly hosts live tech talks and curates articles on big data, recommendation/search algorithms, advertising algorithms, NLP, intelligent risk control, autonomous driving, and machine learning/deep learning.

2.7k
Articles
3
Likes
18.5k
Views
1
Comments
Recent Articles

Latest from DataFunTalk

100 recent articles max
DataFunTalk
DataFunTalk
Oct 7, 2026 · Artificial Intelligence

Ontology-Driven Agent Control: Semantic Foundations for Harness Engineering

The article explains how ontology-driven architecture provides semantic infrastructure for controllable AI Agent execution, detailing three pillars (architectural constraints, context engineering, feedback loops), the Knora platform's layered architecture, and real-world deployments showing 70x efficiency gains.

Agent ControlHarness engineeringKnora Platform
0 likes · 27 min read
Ontology-Driven Agent Control: Semantic Foundations for Harness Engineering
DataFunTalk
DataFunTalk
Oct 7, 2026 · Artificial Intelligence

360's Data Agent: Dual-Path Semantic Layer & Confidence-Driven Evaluation

360's enterprise Data Agent bridges the NL2SQL production gap with a dual-path routing system—AST compilation for mature semantic layers and RAG fallback for sparse metadata—guided by confidence scores that gate automated reporting versus human review, while evolving skills into an ontology-driven knowledge layer and validating via a three-level evaluation flywheel.

Data AgentDual-Path RoutingNL2SQL
0 likes · 6 min read
360's Data Agent: Dual-Path Semantic Layer & Confidence-Driven Evaluation
DataFunTalk
DataFunTalk
Oct 6, 2026 · Artificial Intelligence

AWS Strands Harness Slashes Agent Costs 77% Without Changing Models

AWS open-sourced Strands Harness, an agent runtime that cuts costs from $248 to $56 and boosts accuracy on Terminal-Bench 2.1 by optimizing context management, prompt caching, and layered memory, proving agent performance hinges on the harness not just the model.

AI AgentsAWSAgent Runtime
0 likes · 18 min read
AWS Strands Harness Slashes Agent Costs 77% Without Changing Models
DataFunTalk
DataFunTalk
Oct 5, 2026 · Artificial Intelligence

Why OpenAI Sends Engineers On-Site: The FDE Playbook for Enterprise AI

OpenAI's Forward Deployed Engineering lead Colin Jarvis reveals why stronger models require on-site engineers to bridge the gap between demo capabilities and production trust, detailing eval-driven development, deterministic-probabilistic boundaries, and the 'Eat Pain → Excrete Product' methodology across Morgan Stanley, semiconductor, supply chain, and Klarna case studies.

Agent ArchitectureDeterministic vs Probabilistic SystemsEval-Driven Development
0 likes · 19 min read
Why OpenAI Sends Engineers On-Site: The FDE Playbook for Enterprise AI
DataFunTalk
DataFunTalk
Oct 5, 2026 · Artificial Intelligence

WeChat's Agentic OLAP Architecture: Observability, Memory & Access Control

WeChat's technical architecture team shares their exploration of adapting OLAP infrastructure for Agentic AI, covering distributed observability with Langfuse and ClickHouse, remote memory services using vector search and FUSE, and access protection mechanisms treating agents as authenticated identities.

Agentic AIClickHouseDiskANN
0 likes · 16 min read
WeChat's Agentic OLAP Architecture: Observability, Memory & Access Control
DataFunTalk
DataFunTalk
Oct 4, 2026 · Artificial Intelligence

Why AI Agents Need a Dedicated Decision Layer: Inside Jev's Structured Judgment Model

TypeSafe's Jev model introduces a dedicated Decision Layer for AI agents, replacing free-text generation with structured judgments (Choice, Score, Noul) for high-frequency routing, tool selection, guardrails, and context compaction, enabling faster, cheaper, and more calibrated decisions than full LLM calls.

Agent ArchitectureDecision LayerJev
0 likes · 24 min read
Why AI Agents Need a Dedicated Decision Layer: Inside Jev's Structured Judgment Model
DataFunTalk
DataFunTalk
Oct 3, 2026 · Artificial Intelligence

Palantir's AIP Analyst Adds Skills: Enterprise AI Begins Reusing Analysis Methods

Palantir's August 2024 AIP Analyst update introduces Skills and Analysis Lookup, enabling AI agents to reuse analytical methods and historical analysis templates instead of just retrieving content, adding a method-memory layer atop traditional knowledge memory grounded in Ontology.

AI AgentsAIP AnalystAnalysis Lookup
0 likes · 11 min read
Palantir's AIP Analyst Adds Skills: Enterprise AI Begins Reusing Analysis Methods
DataFunTalk
DataFunTalk
Oct 3, 2026 · Artificial Intelligence

Andrew Ng Redraws AI Engineering Skills: From Coding to Defining & Verifying

DeepLearning.AI analyzed 10,000+ job postings to redefine AI engineering skills into four areas, showing that as coding agents handle implementation, engineers must focus on task definition, architecture, verification, and controlling agent autonomy — illustrated by GitHub's 832k-line Rust migration where humans spent 63% of effort on review, design challenges, and task completion.

AI engineeringAndrew NgCoding Agents
0 likes · 16 min read
Andrew Ng Redraws AI Engineering Skills: From Coding to Defining & Verifying
DataFunTalk
DataFunTalk
Oct 2, 2026 · Artificial Intelligence

Xiaohongshu's GR-Inference: Custom Engine for 3.6x Faster Generative Retrieval

Xiaohongshu built a custom inference engine GR-Inference for generative search retrieval, addressing unique load characteristics — long context, short decode, large dynamic beam, and constrained generation — that break general frameworks, achieving 1.5–3.6x throughput over SGLang and improving recall and click-through rates.

GR-InferenceLLM servingSGLang
0 likes · 7 min read
Xiaohongshu's GR-Inference: Custom Engine for 3.6x Faster Generative Retrieval