Collect first agent research and market sources
This commit is contained in:
wuyang
@@ -6,48 +6,59 @@ status: seed
|
||||
|
||||
## Current Snapshot
|
||||
|
||||
- date:
|
||||
- scope:
|
||||
- sources reviewed:
|
||||
- date: 2026-07-08
|
||||
- scope: large-company Agent work, engineering blogs, benchmarks, product/technical reports
|
||||
- sources reviewed: 9 starter items
|
||||
- source mix: official 9
|
||||
|
||||
## Strong Industry Signals
|
||||
|
||||
-
|
||||
- Enterprise Agent 正在从 demo 转向有权限、记忆、评估、trace、安全和平台入口的系统。
|
||||
- Computer use/GUI Agent 正在成为主流产品能力，但安全控制必须产品化。
|
||||
- Memory 的工业定义正在从“记住信息”转向“通过经验改善流程执行”。
|
||||
- Agent debug/eval 正在围绕 trajectory、constraint、failure taxonomy 和 benchmark 构建。
|
||||
- 大公司开始把 Agent adoption 当成组织生产力指标看，而不是单一模型能力。
|
||||
|
||||
## Weak or Marketing-heavy Signals
|
||||
|
||||
-
|
||||
- 单纯模型发布里提到 “agent frontier” 但缺少 benchmark、代码或系统细节时，只能作为方向信号。
|
||||
- Qwen-AgentWorld 当前抓取到的信息不足，需要补完整正文。
|
||||
- Agent usage growth 指标能说明采用趋势，但不能直接说明任务质量。
|
||||
|
||||
## Company Patterns
|
||||
|
||||
| Company | Agent Direction | Evidence | Notes |
|
||||
| --- | --- | --- | --- |
|
||||
| OpenAI | _TBD_ | _TBD_ | _TBD_ |
|
||||
| Anthropic | _TBD_ | _TBD_ | _TBD_ |
|
||||
| Google DeepMind | _TBD_ | _TBD_ | _TBD_ |
|
||||
| Microsoft | _TBD_ | _TBD_ | _TBD_ |
|
||||
| DeepSeek | _TBD_ | _TBD_ | _TBD_ |
|
||||
| 腾讯 | _TBD_ | _TBD_ | _TBD_ |
|
||||
| 字节 Seed | _TBD_ | _TBD_ | _TBD_ |
|
||||
| 阿里 Qwen | _TBD_ | _TBD_ | _TBD_ |
|
||||
| OpenAI | internal data agent, enterprise adoption, Codex as work agent | in-house data agent, agents transforming work | 系统细节集中在数据 Agent |
|
||||
| Anthropic | computer use and coding agents | not yet itemized | 下一轮补 Claude computer use / Claude Code |
|
||||
| Google DeepMind | computer use, agent security/control | Gemini computer use, AI Control Roadmap | 安全控制非常值得沉淀 |
|
||||
| Microsoft | memory benchmark, failure diagnosis, skill optimization | STATE-Bench, AgentRx, SkillOpt | 评估和工程方法密集 |
|
||||
| DeepSeek | Agent integration ecosystem and hiring wave | jobs item, awesome-deepseek-agent source candidate | 需补官方技术资料 |
|
||||
| 腾讯 | Agent testing / cloud harness signal | jobs item | 需补技术博客和北京岗位 |
|
||||
| 字节 Seed | productivity agent, GUI+MCP, coding agent | Seed2.1 official blog | 与岗位中的 harness/memory/context compression 呼应 |
|
||||
| 阿里 Qwen | AgentWorld / world models for general agents | Qwen-AgentWorld queued item | 需补正文和 Qwen3.x agent blogs |
|
||||
|
||||
## Implementation Patterns
|
||||
|
||||
- tool use:
|
||||
- memory:
|
||||
- evaluation:
|
||||
- safety:
|
||||
- deployment:
|
||||
- productization:
|
||||
- tool use: API/MCP/function calling 与 GUI/computer use 两条线并行。
|
||||
- memory: 从 raw history 转向 structured memory、stateful benchmark、experience consolidation。
|
||||
- evaluation: adoption metrics、benchmark、trajectory failure diagnosis、deterministic assertions。
|
||||
- safety: defense-in-depth、sandboxing、human confirmation、prompt injection guard、permission gating。
|
||||
- deployment: Slack/Web/IDE/CLI/MCP、enterprise platform、internal tools。
|
||||
- productization: coding agent、data agent、office/productivity agent、customer support/travel/shopping workflows。
|
||||
|
||||
## Follow-up Queue
|
||||
|
||||
-
|
||||
- Anthropic Claude computer use / Claude Code / model cards。
|
||||
- OpenAI workspace agents / Frontier / Codex enterprise coding posts。
|
||||
- Qwen AgentWorld and Qwen3.x agent blogs full extraction。
|
||||
- DeepSeek official GitHub agent ecosystem and talent site details。
|
||||
- Tencent AI Lab / Hunyuan / Yuanbao Agent technical posts。
|
||||
|
||||
## Links Back to Knowledge Base
|
||||
|
||||
- docs to update:
|
||||
- papers to read:
|
||||
- jobs to compare:
|
||||
- experiments to create:
|
||||
- projects affected:
|
||||
- docs to update: evaluation, observability, memory, safety, practice playbook
|
||||
- papers to read: OpenAgentSafety, memory survey, EvoMemBench, SWE-EVO, Dialogue-SWEBench
|
||||
- jobs to compare: Baidu Agent algorithm/fullstack, DeepSeek Agent hiring, Tencent Agent testing, ByteDance Seed Agent research
|
||||
- experiments to create: memory benchmark, trace/failure diagnosis, GUI+MCP workflow
|
||||
- projects affected: future knowledge frontend, coding-agent experiments, eval harness
|
||||
|
||||
@@ -0,0 +1,72 @@
|
||||
# Industry: ByteDance Seed - Seed2.1 Agent Productivity
|
||||
|
||||
---
|
||||
type: industry
|
||||
company: ByteDance Seed
|
||||
team: Seed
|
||||
title: "Seed2.1 Officially Released: Advancing AI Productivity"
|
||||
url: https://seed.bytedance.com/en/blog/seed2-1-officially-released-advancing-ai-productivity
|
||||
source_name: ByteDance Seed
|
||||
source_type: research-blog
|
||||
source_quality: official
|
||||
published_at:
|
||||
collected_at: 2026-07-08
|
||||
status: analyzed
|
||||
topics:
|
||||
- agent
|
||||
- coding-agent
|
||||
- gui-agent
|
||||
implementation_signals:
|
||||
- reinforcement-learning
|
||||
- gui-tool-use
|
||||
- mcp
|
||||
- crowdsourced-evaluation
|
||||
- internal-benchmark
|
||||
product_area:
|
||||
- productivity
|
||||
- coding-agent
|
||||
- office-agent
|
||||
models:
|
||||
- Seed2.1
|
||||
tools:
|
||||
- MCP
|
||||
benchmarks:
|
||||
- CreativeWork
|
||||
- ProgramBench
|
||||
related_papers:
|
||||
- 2026-pham-swe-evo
|
||||
related_jobs:
|
||||
- 2026-07-08-bytedance-seed-llm-agent-research-engineer
|
||||
related_experiments:
|
||||
-
|
||||
related_projects:
|
||||
-
|
||||
evidence_level: high
|
||||
relevance: high
|
||||
---
|
||||
|
||||
## One-line Takeaway
|
||||
|
||||
ByteDance Seed 把 Agent 能力放在办公生产力、GUI+MCP 工具协作和端到端 coding delivery 三个方向同时推进。
|
||||
|
||||
## What They Built or Claimed
|
||||
|
||||
Seed2.1 强调通用 Agent 和代码工程能力，使用 RL 提升 GUI/非 GUI 行动选择效率，并提出 CreativeWork 内部 benchmark 评估 Notion、Canva、Figma 等生产力场景中的 GUI 与 MCP 协同。
|
||||
|
||||
## Technical Signals
|
||||
|
||||
- architecture: model + GUI action space + MCP tools。
|
||||
- tool use: GUI 与 MCP 工具协调。
|
||||
- memory:
|
||||
- evaluation: CreativeWork、ProgramBench、众包开发者反馈、内部评估。
|
||||
- safety:
|
||||
- deployment: productivity 和 coding scenarios。
|
||||
- data: 真实代码仓库众包任务反馈。
|
||||
|
||||
## Evidence Quality
|
||||
|
||||
官方 Seed 博客，包含 benchmark、场景和评估方式，证据强；具体数据表需后续截图/原文复核。
|
||||
|
||||
## Gaps for Us
|
||||
|
||||
- 需要研究 GUI+MCP 的 Agent 工作流设计。
|
||||
@@ -0,0 +1,70 @@
|
||||
# Industry: Google DeepMind - AI Control Roadmap for Agents
|
||||
|
||||
---
|
||||
type: industry
|
||||
company: Google DeepMind
|
||||
team:
|
||||
title: Securing the future of AI agents
|
||||
url: https://deepmind.google/blog/securing-the-future-of-ai-agents/
|
||||
source_name: Google DeepMind
|
||||
source_type: technical-report
|
||||
source_quality: official
|
||||
published_at:
|
||||
collected_at: 2026-07-08
|
||||
status: analyzed
|
||||
topics:
|
||||
- agent-safety
|
||||
- agent-security
|
||||
- governance
|
||||
implementation_signals:
|
||||
- defense-in-depth
|
||||
- sandboxing
|
||||
- permission-control
|
||||
- prompt-injection-resistance
|
||||
product_area:
|
||||
- internal-agents
|
||||
- enterprise-agent
|
||||
models:
|
||||
-
|
||||
tools:
|
||||
-
|
||||
benchmarks:
|
||||
-
|
||||
related_papers:
|
||||
- 2025-vijayvargiya-openagentsafety
|
||||
- 2026-agent-safety-benchmark-taxonomy
|
||||
related_jobs:
|
||||
- 2026-07-08-tencent-cloud-ai-agent-test-engineer
|
||||
related_experiments:
|
||||
-
|
||||
related_projects:
|
||||
-
|
||||
evidence_level: high
|
||||
relevance: high
|
||||
---
|
||||
|
||||
## One-line Takeaway
|
||||
|
||||
Agent 安全需要把内部 Agent 当作潜在 insider threat 来设计权限、沙箱和逐步授权。
|
||||
|
||||
## What They Built or Claimed
|
||||
|
||||
Google DeepMind 介绍 AI Control Roadmap，用 defense-in-depth 保护 Google 内部系统，应对能力增强但不完美对齐的 Agent。
|
||||
|
||||
## Technical Signals
|
||||
|
||||
- architecture: alignment + system-level AI control + cybersecurity-style threat modeling。
|
||||
- tool use:
|
||||
- memory:
|
||||
- evaluation: 通过行为验证逐步授予权限。
|
||||
- safety: sandboxing、endpoint security、prompt injection resistance、权限分层。
|
||||
- deployment: 内部 Agent 系统治理。
|
||||
- data:
|
||||
|
||||
## Evidence Quality
|
||||
|
||||
官方技术/治理材料，适合作为 Agent 安全架构参考。
|
||||
|
||||
## Gaps for Us
|
||||
|
||||
- 需要把 Agent 权限分层写入实践 playbook。
|
||||
@@ -0,0 +1,70 @@
|
||||
# Industry: Google DeepMind - Computer Use in Gemini 3.5 Flash
|
||||
|
||||
---
|
||||
type: industry
|
||||
company: Google DeepMind
|
||||
team: Gemini
|
||||
title: Introducing computer use in Gemini 3.5 Flash
|
||||
url: https://deepmind.google/blog/introducing-computer-use-in-gemini-3-5-flash/
|
||||
source_name: Google DeepMind / Google Blog
|
||||
source_type: product-blog
|
||||
source_quality: official
|
||||
published_at: 2026-06-24
|
||||
collected_at: 2026-07-08
|
||||
status: analyzed
|
||||
topics:
|
||||
- computer-use
|
||||
- agent
|
||||
- tool-use
|
||||
implementation_signals:
|
||||
- browser-control
|
||||
- gui-control
|
||||
- safety
|
||||
- human-in-the-loop
|
||||
product_area:
|
||||
- enterprise-agent
|
||||
- automation
|
||||
models:
|
||||
- Gemini 3.5 Flash
|
||||
tools:
|
||||
- Gemini API
|
||||
- Gemini Enterprise Agent Platform
|
||||
benchmarks:
|
||||
-
|
||||
related_papers:
|
||||
- 2025-vijayvargiya-openagentsafety
|
||||
related_jobs:
|
||||
- 2026-07-08-baidu-aidu-agent-fullstack-engineer-beijing
|
||||
related_experiments:
|
||||
-
|
||||
related_projects:
|
||||
-
|
||||
evidence_level: high
|
||||
relevance: high
|
||||
---
|
||||
|
||||
## One-line Takeaway
|
||||
|
||||
Computer use 正在从独立模型能力并入主模型和企业 Agent 平台，安全控制成为产品能力的一部分。
|
||||
|
||||
## What They Built or Claimed
|
||||
|
||||
Google 表示 Gemini 3.5 Flash 原生支持 computer use，可让 Agent 在浏览器、移动端和桌面环境中看、推理并行动，支持长程和企业自动化任务。
|
||||
|
||||
## Technical Signals
|
||||
|
||||
- architecture: computer use 作为 Gemini 3.5 Flash 内置工具能力。
|
||||
- tool use: browser/mobile/desktop GUI actions。
|
||||
- memory:
|
||||
- evaluation: 提到 agentic computer use tasks，但需要进一步追具体 benchmark。
|
||||
- safety: 对 computer use 做 adversarial training；企业 safeguard 支持敏感/不可逆动作确认和间接 prompt injection 自动停止。
|
||||
- deployment: Gemini API 和 Gemini Enterprise Agent Platform。
|
||||
- data:
|
||||
|
||||
## Evidence Quality
|
||||
|
||||
官方产品/技术博客，有明确能力边界、安全控制和平台入口。
|
||||
|
||||
## Gaps for Us
|
||||
|
||||
- 需要整理 computer-use Agent 的安全 checklist。
|
||||
@@ -0,0 +1,68 @@
|
||||
# Industry: Microsoft Research - AgentRx
|
||||
|
||||
---
|
||||
type: industry
|
||||
company: Microsoft
|
||||
team: Microsoft Research
|
||||
title: "Systematic debugging for AI agents: Introducing the AgentRx framework"
|
||||
url: https://www.microsoft.com/en-us/research/blog/systematic-debugging-for-ai-agents-introducing-the-agentrx-framework/
|
||||
source_name: Microsoft Research
|
||||
source_type: research-blog
|
||||
source_quality: official
|
||||
published_at:
|
||||
collected_at: 2026-07-08
|
||||
status: analyzed
|
||||
topics:
|
||||
- agent-evaluation
|
||||
- observability
|
||||
- debugging
|
||||
implementation_signals:
|
||||
- trajectory-normalization
|
||||
- constraint-synthesis
|
||||
- constraint-checking
|
||||
- failure-taxonomy
|
||||
product_area:
|
||||
- agent-debugging
|
||||
models:
|
||||
-
|
||||
tools:
|
||||
- AgentRx
|
||||
benchmarks:
|
||||
- AgentRx Benchmark
|
||||
related_papers:
|
||||
- 2025-vijayvargiya-openagentsafety
|
||||
related_jobs:
|
||||
- 2026-07-08-tencent-cloud-ai-agent-test-engineer
|
||||
related_experiments:
|
||||
-
|
||||
related_projects:
|
||||
-
|
||||
evidence_level: high
|
||||
relevance: high
|
||||
---
|
||||
|
||||
## One-line Takeaway
|
||||
|
||||
Agent 调试需要从“最终失败”定位到 trajectory 中第一个不可恢复错误步骤。
|
||||
|
||||
## What They Built or Claimed
|
||||
|
||||
Microsoft Research 发布 AgentRx 和 115 条人工标注失败轨迹 benchmark，用于定位 Agent trajectory 的 critical failure step。
|
||||
|
||||
## Technical Signals
|
||||
|
||||
- architecture: trajectory normalization -> constraint synthesis -> guarded evaluation -> LLM judge。
|
||||
- tool use: 根据 tool schema 和 domain policy 生成可执行约束。
|
||||
- memory:
|
||||
- evaluation: 逐步检查约束，输出可审计验证日志。
|
||||
- safety: 区分计划偏离、幻觉、新信息发明等失败类型。
|
||||
- deployment: domain-agnostic debugging framework。
|
||||
- data: tau-bench、Flash、Magentic-One 失败轨迹。
|
||||
|
||||
## Evidence Quality
|
||||
|
||||
官方研究博客，包含框架、benchmark 规模、失败分类，证据强。
|
||||
|
||||
## Gaps for Us
|
||||
|
||||
- 本知识库需要设计统一 Agent trace schema 和失败 taxonomy。
|
||||
@@ -0,0 +1,68 @@
|
||||
# Industry: Microsoft Research - SkillOpt
|
||||
|
||||
---
|
||||
type: industry
|
||||
company: Microsoft
|
||||
team: Microsoft Research
|
||||
title: "SkillOpt: Agent skills as trainable parameters"
|
||||
url: https://www.microsoft.com/en-us/research/blog/skillopt-agent-skills-as-trainable-parameters/
|
||||
source_name: Microsoft Research
|
||||
source_type: research-blog
|
||||
source_quality: official
|
||||
published_at: 2026-06-30
|
||||
collected_at: 2026-07-08
|
||||
status: analyzed
|
||||
topics:
|
||||
- agent
|
||||
- prompt-optimization
|
||||
- skill-learning
|
||||
implementation_signals:
|
||||
- skill-files
|
||||
- validation-gating
|
||||
- bounded-edits
|
||||
- eval-loop
|
||||
product_area:
|
||||
- agent-framework
|
||||
models:
|
||||
-
|
||||
tools:
|
||||
-
|
||||
benchmarks:
|
||||
- six-benchmark-evaluation
|
||||
related_papers:
|
||||
-
|
||||
related_jobs:
|
||||
- 2026-07-08-deepseek-agent-hiring-wave-beijing-hangzhou
|
||||
related_experiments:
|
||||
-
|
||||
related_projects:
|
||||
-
|
||||
evidence_level: high
|
||||
relevance: high
|
||||
---
|
||||
|
||||
## One-line Takeaway
|
||||
|
||||
Agent skill/prompt 文件可以被当成外部可训练参数，用评估循环优化，而不是靠人工随手改。
|
||||
|
||||
## What They Built or Claimed
|
||||
|
||||
Microsoft Research 提出 SkillOpt，将 skill editing 转成训练过程，在不改模型权重的情况下提升 Agent 行为可靠性。
|
||||
|
||||
## Technical Signals
|
||||
|
||||
- architecture: frozen target model + external skill file optimization。
|
||||
- tool use:
|
||||
- memory:
|
||||
- evaluation: 六个 benchmark、七个目标模型、三种执行模式。
|
||||
- safety: bounded text edits、validation gating、rejected-edit feedback、slow/meta updates。
|
||||
- deployment: 可审计技能文件。
|
||||
- data:
|
||||
|
||||
## Evidence Quality
|
||||
|
||||
官方研究博客给出实验覆盖和机制细节，证据强。
|
||||
|
||||
## Gaps for Us
|
||||
|
||||
- 可以把本项目的 `skills`、prompt、playbook 也纳入 eval-driven 优化。
|
||||
@@ -0,0 +1,71 @@
|
||||
# Industry: Microsoft - STATE-Bench
|
||||
|
||||
---
|
||||
type: industry
|
||||
company: Microsoft
|
||||
team: Microsoft Open Source
|
||||
title: "Introducing STATE-Bench: a benchmark for AI agent memory"
|
||||
url: https://opensource.microsoft.com/blog/2026/05/19/introducing-state-bench-a-benchmark-for-ai-agent-memory/
|
||||
source_name: Microsoft Open Source Blog
|
||||
source_type: benchmark
|
||||
source_quality: official
|
||||
published_at: 2026-05-19
|
||||
collected_at: 2026-07-08
|
||||
status: analyzed
|
||||
topics:
|
||||
- memory
|
||||
- agent-evaluation
|
||||
- enterprise-ai
|
||||
implementation_signals:
|
||||
- stateful-environment
|
||||
- user-simulator
|
||||
- deterministic-assertions
|
||||
- bring-your-own-memory
|
||||
product_area:
|
||||
- customer-support
|
||||
- travel
|
||||
- shopping
|
||||
models:
|
||||
-
|
||||
tools:
|
||||
- STATE-Bench
|
||||
benchmarks:
|
||||
- STATE-Bench
|
||||
related_papers:
|
||||
- 2026-memory-agent-survey
|
||||
- 2026-wang-evomembench
|
||||
related_jobs:
|
||||
- 2026-07-08-baidu-aidu-agent-algorithm-engineer-beijing
|
||||
related_experiments:
|
||||
-
|
||||
related_projects:
|
||||
-
|
||||
evidence_level: high
|
||||
relevance: high
|
||||
---
|
||||
|
||||
## One-line Takeaway
|
||||
|
||||
Memory 评估应衡量 Agent 是否因经验而更会执行流程，而不是只测试远距离事实召回。
|
||||
|
||||
## What They Built or Claimed
|
||||
|
||||
Microsoft 发布 STATE-Bench，一个开源、memory-agnostic benchmark，用现实企业任务评估 Agent 是否能通过经验改进。
|
||||
|
||||
## Technical Signals
|
||||
|
||||
- architecture: 多轮对话 loop + stateful environment + deterministic assertions。
|
||||
- tool use: 任务涉及查询、校验策略、计算费用、确认并执行。
|
||||
- memory: bring-your-own-memory 接口。
|
||||
- evaluation: 450 个任务，覆盖 travel、customer support、shopping。
|
||||
- safety: 状态改变任务会带来真实成本，强调流程合规。
|
||||
- deployment: open-source benchmark。
|
||||
- data: 预填充数据库和用户模拟器。
|
||||
|
||||
## Evidence Quality
|
||||
|
||||
官方开源博客，包含任务数、领域、评估 loop 和 GitHub 入口，证据强。
|
||||
|
||||
## Gaps for Us
|
||||
|
||||
- 可以基于 STATE-Bench 思路做一个小型企业流程 Agent eval。
|
||||
@@ -0,0 +1,75 @@
|
||||
# Industry: OpenAI - How Agents Are Transforming Work
|
||||
|
||||
---
|
||||
type: industry
|
||||
company: OpenAI
|
||||
team:
|
||||
title: How agents are transforming work
|
||||
url: https://openai.com/index/how-agents-are-transforming-work/
|
||||
source_name: OpenAI
|
||||
source_type: research-blog
|
||||
source_quality: official
|
||||
published_at:
|
||||
collected_at: 2026-07-08
|
||||
status: analyzed
|
||||
topics:
|
||||
- agent
|
||||
- enterprise-ai
|
||||
- coding-agent
|
||||
implementation_signals:
|
||||
- adoption-metrics
|
||||
- workflow-change
|
||||
product_area:
|
||||
- coding-agent
|
||||
- enterprise-work
|
||||
models:
|
||||
- Codex
|
||||
tools:
|
||||
-
|
||||
benchmarks:
|
||||
-
|
||||
related_papers:
|
||||
-
|
||||
related_jobs:
|
||||
- 2026-07-08-deepseek-agent-hiring-wave-beijing-hangzhou
|
||||
related_experiments:
|
||||
-
|
||||
related_projects:
|
||||
-
|
||||
evidence_level: medium
|
||||
relevance: high
|
||||
---
|
||||
|
||||
## One-line Takeaway
|
||||
|
||||
Agent 采用正在从“聊天辅助”变成“可计量的 agentic labor”，企业内部不同职能的增长速度差异值得追踪。
|
||||
|
||||
## What They Built or Claimed
|
||||
|
||||
OpenAI 公开了内部 Codex 使用变化，显示过去六个月内研究、客服、工程、法务等部门的 agentic usage 明显增长，并认为员工正在从 chatbot 转向 agent 作为主要 AI 交互形态。
|
||||
|
||||
## Technical Signals
|
||||
|
||||
- architecture: 未详细公开。
|
||||
- tool use: Codex 作为工作代理。
|
||||
- memory:
|
||||
- evaluation: 使用 adoption / output token 增长作为组织层面的采用指标。
|
||||
- safety:
|
||||
- deployment: 企业内部多部门使用。
|
||||
- data: 部门级使用增长。
|
||||
|
||||
## Evidence Quality
|
||||
|
||||
官方数据型博客，适合看采用趋势；技术实现细节较少。
|
||||
|
||||
## Links
|
||||
|
||||
- papers:
|
||||
- jobs: DeepSeek Agent hiring wave
|
||||
- experiments:
|
||||
- projects:
|
||||
- docs: practice playbook
|
||||
|
||||
## Gaps for Us
|
||||
|
||||
- 需要设计自己的 Agent 使用指标：任务数、回合数、人工改动量、成功率、耗时和成本。
|
||||
@@ -0,0 +1,81 @@
|
||||
# Industry: OpenAI - In-house Data Agent
|
||||
|
||||
---
|
||||
type: industry
|
||||
company: OpenAI
|
||||
team: data / engineering
|
||||
title: Inside OpenAI's in-house data agent
|
||||
url: https://openai.com/index/inside-our-in-house-data-agent/
|
||||
source_name: OpenAI
|
||||
source_type: engineering-blog
|
||||
source_quality: official
|
||||
published_at:
|
||||
collected_at: 2026-07-08
|
||||
status: analyzed
|
||||
topics:
|
||||
- agent
|
||||
- data-agent
|
||||
- enterprise-ai
|
||||
implementation_signals:
|
||||
- memory
|
||||
- rag
|
||||
- mcp
|
||||
- eval
|
||||
- permissions
|
||||
product_area:
|
||||
- data-analysis
|
||||
- internal-tools
|
||||
models:
|
||||
- GPT-5.2
|
||||
- Codex
|
||||
tools:
|
||||
- Evals API
|
||||
- Embeddings API
|
||||
- MCP
|
||||
benchmarks:
|
||||
-
|
||||
related_papers:
|
||||
- 2026-memory-agent-survey
|
||||
related_jobs:
|
||||
- 2026-07-08-baidu-aidu-agent-fullstack-engineer-beijing
|
||||
related_experiments:
|
||||
-
|
||||
related_projects:
|
||||
-
|
||||
evidence_level: high
|
||||
relevance: high
|
||||
---
|
||||
|
||||
## One-line Takeaway
|
||||
|
||||
高价值内部 Agent 不是只接一个数据库，而是把权限、表级知识、人工注释、代码增强、组织知识、记忆和运行时上下文做成系统。
|
||||
|
||||
## What They Built or Claimed
|
||||
|
||||
OpenAI 构建了内部数据 Agent，用自然语言帮助员工探索和分析公司数据平台。资料描述了多层上下文、Agent API、Slack/Web/CLI/MCP 入口、记忆系统和安全边界。
|
||||
|
||||
## Technical Signals
|
||||
|
||||
- architecture: Agent API + internal knowledge + warehouse/platform sources + model/tool loop。
|
||||
- tool use: 查询数据、语义检索、精确文本检索、MCP 连接。
|
||||
- memory: 持续学习的记忆系统，用于提升后续查询。
|
||||
- evaluation: 使用 Evals API 作为构建工具之一。
|
||||
- safety: 内部权限和工作流是设计的一部分。
|
||||
- deployment: Slack、Web、IDE、Codex CLI、内部 ChatGPT。
|
||||
- data: 表使用、人工注释、Codex enrichment、institutional knowledge、runtime context。
|
||||
|
||||
## Evidence Quality
|
||||
|
||||
官方工程博客，有系统结构、上下文层和落地入口，证据强。
|
||||
|
||||
## Links
|
||||
|
||||
- papers: memory survey
|
||||
- jobs: Baidu Agent fullstack
|
||||
- experiments: data-agent-context-stack
|
||||
- projects:
|
||||
- docs: evaluation, memory, tool use
|
||||
|
||||
## Gaps for Us
|
||||
|
||||
- 需要沉淀一个“企业数据 Agent 架构”专题。
|
||||
@@ -0,0 +1,67 @@
|
||||
# Industry: Qwen - AgentWorld
|
||||
|
||||
---
|
||||
type: industry
|
||||
company: Alibaba Qwen
|
||||
team: Qwen
|
||||
title: "Qwen-AgentWorld: Language World Models for General Agents"
|
||||
url: https://qwen.ai/blog?id=qwen-agentworld
|
||||
source_name: Qwen Blog
|
||||
source_type: research-blog
|
||||
source_quality: official
|
||||
published_at:
|
||||
collected_at: 2026-07-08
|
||||
status: queued
|
||||
topics:
|
||||
- agent
|
||||
- world-model
|
||||
- evaluation
|
||||
implementation_signals:
|
||||
- simulated-environment
|
||||
- world-model
|
||||
- multi-domain-agent
|
||||
product_area:
|
||||
- general-agent
|
||||
models:
|
||||
- Qwen
|
||||
tools:
|
||||
-
|
||||
benchmarks:
|
||||
-
|
||||
related_papers:
|
||||
-
|
||||
related_jobs:
|
||||
-
|
||||
related_experiments:
|
||||
-
|
||||
related_projects:
|
||||
-
|
||||
evidence_level: medium
|
||||
relevance: high
|
||||
---
|
||||
|
||||
## One-line Takeaway
|
||||
|
||||
Qwen 正在把通用 Agent 能力和“语言世界模型/环境模拟”联系起来，值得后续重点读完整技术细节。
|
||||
|
||||
## What They Built or Claimed
|
||||
|
||||
搜索摘要显示 Qwen 发布 Qwen-AgentWorld，一个 native language world model，用于模拟七个领域的 Agent 环境。
|
||||
|
||||
## Technical Signals
|
||||
|
||||
- architecture: language world model for simulated agent environments。
|
||||
- tool use:
|
||||
- memory:
|
||||
- evaluation: 多领域环境模拟可能用于 Agent 训练或评估。
|
||||
- safety:
|
||||
- deployment:
|
||||
- data:
|
||||
|
||||
## Evidence Quality
|
||||
|
||||
官方博客入口存在，但当前抓取未展开正文；先作为 queued industry item，后续需要浏览器或脚本补全。
|
||||
|
||||
## Gaps for Us
|
||||
|
||||
- 需要详细阅读 Qwen-AgentWorld，判断它是训练环境、评估环境还是产品化组件。
|
||||
Reference in New Issue
Block a user