Agentic Search
Overview
Agentic search is retrieval where an AI system actively plans, queries, follows leads, compares sources, and decides when it has enough evidence. It goes beyond one-shot RAG by treating search as an iterative reasoning and tool-use process.
Conference Context
It combines web search, enterprise search, information retrieval, RAG, semantic search, BM25, vector databases, knowledge graphs, and research-agent workflows. Agents add query reformulation, source triage, multi-hop exploration, and evidence synthesis.
Significance
Many tasks fail because the agent either retrieves the wrong context or stops too early. Agentic search improves coverage, reduces hallucination, and helps systems expose the evidence behind an answer.
Applied Use
Define the question, retrieve broadly, rerank by task relevance, inspect primary sources, track claims and citations, and loop when evidence conflicts or gaps remain. Use hybrid retrieval and structured indexes where pure vector search misses exact terms or relationships.
It is useful in research, support knowledge bases, compliance review, code search, enterprise assistants, competitive intelligence, and document-heavy operations.
Use agentic search when answers require multiple sources, fresh evidence, exact facts, or cross-document reasoning. Simple lookup or direct database queries are better for narrow deterministic questions.
Define the question, retrieve broadly, rerank by task relevance, inspect primary sources, track claims and citations, and loop when evidence conflicts or gaps remain. Use hybrid retrieval and structured indexes where pure vector search misses exact terms or relationships.
It is useful in research, support knowledge bases, compliance review, code search, enterprise assistants, competitive intelligence, and document-heavy operations.
Use agentic search when answers require multiple sources, fresh evidence, exact facts, or cross-document reasoning. Simple lookup or direct database queries are better for narrow deterministic questions.
Define the question, retrieve broadly, rerank by task relevance, inspect primary sources, track claims and citations, and loop when evidence conflicts or gaps remain. Use hybrid retrieval and structured indexes where pure vector search misses exact terms or relationships.
It is useful in research, support knowledge bases, compliance review, code search, enterprise assistants, competitive intelligence, and document-heavy operations.
Use agentic search when answers require multiple sources, fresh evidence, exact facts, or cross-document reasoning. Simple lookup or direct database queries are better for narrow deterministic questions.
Define the question, retrieve broadly, rerank by task relevance, inspect primary sources, track claims and citations, and loop when evidence conflicts or gaps remain. Use hybrid retrieval and structured indexes where pure vector search misses exact terms or relationships.
It is useful in research, support knowledge bases, compliance review, code search, enterprise assistants, competitive intelligence, and document-heavy operations.
Use agentic search when answers require multiple sources, fresh evidence, exact facts, or cross-document reasoning. Simple lookup or direct database queries are better for narrow deterministic questions.
Define the question, retrieve broadly, rerank by task relevance, inspect primary sources, track claims and citations, and loop when evidence conflicts or gaps remain. Use hybrid retrieval and structured indexes where pure vector search misses exact terms or relationships.
It is useful in research, support knowledge bases, compliance review, code search, enterprise assistants, competitive intelligence, and document-heavy operations.
Use agentic search when answers require multiple sources, fresh evidence, exact facts, or cross-document reasoning. Simple lookup or direct database queries are better for narrow deterministic questions.
Define the question, retrieve broadly, rerank by task relevance, inspect primary sources, track claims and citations, and loop when evidence conflicts or gaps remain. Use hybrid retrieval and structured indexes where pure vector search misses exact terms or relationships.
It is useful in research, support knowledge bases, compliance review, code search, enterprise assistants, competitive intelligence, and document-heavy operations.
Use agentic search when answers require multiple sources, fresh evidence, exact facts, or cross-document reasoning. Simple lookup or direct database queries are better for narrow deterministic questions.
Connections
- 2026 07 01 session vector isn t enough hybrid search and retrieval for ai engineers — Vector Isn't Enough: Hybrid Search & Retrieval for AI Engineers; Jeff Vestal (Day 1 — Workshop Day · 2:20pm-4:20pm · Track 7; official schedule)
- 2026 06 29 jo kristian bergum the unreasonable effectiveness of bm25 for agentic search — The unreasonable effectiveness of BM25 for agentic search; Jo Kristian Bergum (Day 2 — Session Day 1 · 11:10am-11:30am · Search & Retrieval; official schedule)
- 2026 06 29 will bryk the search engine for the agentic web — The Search Engine for the Agentic Web; Will Bryk (Day 2 — Session Day 1 · 11:40am-12:00pm · Search & Retrieval; official schedule)
- 2026 06 29 maximilian david rumpf where rl will take search — Where RL Will Take Search; Maximilian-David Rumpf, Lotte Seifert (Day 2 — Session Day 1 · 2:50pm-3:10pm · Search & Retrieval; official schedule)
- 2026 06 30 han xiao autoresearch for dense retrieval test time compute with frozen embedding models — Autoresearch for Dense Retrieval: Test-Time Compute with Frozen Embedding Models; Han Xiao (Day 3 — Session Day 2 · 11:10am-11:30am · Autoresearch; official schedule)
- 2026 06 30 elie bakouch the era of auto research — « the era of (auto) research »; Elie Bakouch (Day 3 — Session Day 2 · 12:05pm-12:25pm · Autoresearch; official schedule)
- 2026 06 30 tim sweeney closing the loop an autonomous ai research agent — Closing the Loop: An Autonomous AI Research Agent; Tim Sweeney (Day 3 — Session Day 2 · 1:30pm-1:50pm · Autoresearch; official schedule)
- 2026 06 29 dhruv nathawani teaching agents to search building synthetic training pipelines with nvidia data designer — Teaching Agents to Search: Building Synthetic Training Pipelines with NVIDIA Data Designer; Dhruv Nathawani (Day 1 — Workshop Day · 11:05am-12:05pm · Workshops Day 1; official schedule)
- 2026 06 30 erina karati autoresearch in a multi agent ai village — Autoresearch in a Multi-Agent AI Village; Erina Karati, Arunachalam Manikandan (Day 3 — Session Day 2 · 3:45pm-4:05pm · Autoresearch; official schedule)
- 2026 07 01 stephen chin crabrag why automated assistants need graph memory not more tokens — CrabRAG: Why Automated Assistants Need Graph Memory, Not More Tokens; Stephen Chin (Day 4 — Session Day 3 · 10:45am-11:05am · Graphs; official schedule)
- 2026 06 29 zhengyao jiang hands on autoresearch cracking openai s parameter golf — Hands-on AutoResearch: Cracking OpenAI's Parameter Golf; Zhengyao Jiang, Dixing Xu, Vayum Arora, Dhruv Srikanth (Day 1 — Workshop Day · 2:20pm-4:20pm · Workshops Day 1; official schedule)
- 2026 06 30 benoit schillings research to reality with google deepmind — Research to Reality with Google DeepMind; Benoit Schillings (Day 3 — Session Day 2 · 10:05am-10:25am · Autoresearch; official schedule)
- 2026 06 30 richard socher first steps toward automated ai research — First Steps Toward Automated AI Research; Richard Socher (Day 3 — Session Day 2 · 10:45am-11:05am · Autoresearch; official schedule)
- 2026 06 30 tejas bhakta autoresearch for kernels — Autoresearch for Kernels; Tejas Bhakta (Day 3 — Session Day 2 · 2:50pm-3:10pm · Autoresearch; official schedule)
- 2026 06 30 roland gavrilescu autoresearch in the wild — Autoresearch in the wild; Roland Gavrilescu, Julian Bright (Day 3 — Session Day 2 · 3:20pm-3:40pm · Autoresearch; official schedule)
- 2026 07 01 brendan rappazzo alphalab autonomous multi agent research across optimization domains with frontier llms — ALPHALAB: Autonomous Multi-Agent Research Across Optimization Domains with Frontier LLMs; Brendan Rappazzo (Day 4 — Session Day 3 · 10:45am-11:05am · AI in Finance; official schedule)
- 2026 06 29 nyah macklin rag needs a map using graphrag to retrieve connected context — RAG Needs a Map: Using GraphRAG to Retrieve Connected Context; Nyah Macklin (Day 1 — Workshop Day · 11:05am-12:05pm · Track 2; official schedule)
- 2026 06 29 jess wang agentic vs vector search an eval driven approach to coding agent performance — Agentic vs. Vector Search: An Eval-Driven Approach to Coding Agent Performance; Jess Wang (Day 2 — Session Day 1 · 11:40am-12:00pm · Expo Stage 2 NW; official schedule)
- 2026 06 30 nixon dinh the death of keyword search and the rise of agent readable catalogs — The Death of Keyword Search and the Rise of Agent-Readable Catalogs; Nixon Dinh (Day 3 — Session Day 2 · 11:10am-11:30am · Expo Stage 3; official schedule)
- 2026 07 01 george he everyone talks about document search but what about results — Everyone talks about document search, but what about results?; George He (Day 4 — Session Day 3 · 1:55pm-2:15pm · Expo Stage 4 SE; official schedule)
- 2026 06 30 stefania druga memory harnesses for long running research agents — Memory Harnesses for Long-Running Research Agents; Stefania Druga (Day 3 — Session Day 2 · 11:40am-12:00pm · Memory & Continual Learning; official schedule)
- 2026 07 01 zubin aysola aria how we built autoresearch with autoresearch — ARIA, how we built autoresearch with autoresearch; Zubin Aysola (Day 4 — Session Day 3 · 11:10am-11:30am · Expo Stage 2 NW; official schedule)
- 2026 06 29 peter werry beyond rag build a relational context engine from scratch — Beyond RAG: Build a Relational Context Engine from Scratch; Peter Werry (Day 1 — Workshop Day · 12:10pm-1:10pm · Workshops Day 1; official schedule)
- 2026 06 29 valeria wu fon speech to speech model research at google deepmind — Speech-to-Speech Model Research at Google DeepMind; Valeria Wu Fon, Tom Ouyang (Day 2 — Session Day 1 · 11:10am-11:30am · Voice & Realtime AI; official schedule)
- Zhengyao Jiang
- Brandon Waselnuk
- Kent C. Dodds
- Abhishek Bhardwaj
- Jeff Vestal
- Jo Kristian Bergum
- Will Bryk
- Maximilian-David Rumpf
- Lotte Seifert
- Han Xiao
- Elie Bakouch
- Tim Sweeney
- Dhruv Nathawani
- Erina Karati
- Arunachalam Manikandan
- Stephen Chin
- Dixing Xu
- Vayum Arora
- Dhruv Srikanth
- Benoit Schillings
- Richard Socher
- Tejas Bhakta
- Roland Gavrilescu
- Julian Bright
- NVIDIA
- Weco AI
- Google DeepMind
- Neo4j
- Unblocked
- Bright Data
- Oracle
- Amazon AGI Lab
- Elastic
- Weights & Biases by CoreWeave
- Introspection
- Exa
- turbopuffer
- LlamaIndex
- Artificial Analysis
- Prime Intellect
- DatologyAI
- Browserbase
- youtube HsxQICTLF84 slides — Building an ACP-Compatible Agent Live — Bennet Fenner, Zed (5 extracted slide frames)
- youtube IQkVMvXQKLY slides — Your LLM Deception Monitor Is Broken. The Fix Is in the Training Data - Sachin Kumar, LexisNexis (14 extracted slide frames)
- youtube 1IdzkRVmWAA slides — How we taught agents to use good retrieval - Hanna Lichtenberg, Mixedbread AI (5 extracted slide frames)
- youtube 2e9ANoOEn28 slides — What if the harness mattered more than the model? - Aditya Bhargava, Etsy (8 extracted slide frames)
- youtube CLttOU7n6sI slides — Respect The Process - Andrew Dumit, Watershed Technology Inc. (16 extracted slide frames)
- youtube UcYoMg 8 L8 slides — 500 people vibe-coded for 30 days. I was one of them. - Sanja Grbic, Automattic (11 extracted slide frames)
- youtube 2IxD9OB3XuQ slides — Continual Learning for AI Agents: From Failures to Durable Improvements - Soheil Feizi, RELAI (24 extracted slide frames)
Evidence Graph
Transcript-backed resources
- youtube htM02KMNZnk — WF2026: Software Factories & Keynotes ft. Microsoft, OpenAI, OpenClaw, Z.ai (GLM), MiniMax, HF
- youtube 9fubhllmsBU — Field Guide to Fable — Thariq Shihipar, Anthropic
- youtube 8G_1 3IO4ZQ — WTF Is the Context Layer? The Missing Infrastructure for Production Agents — Prukalpa Sankar
- youtube WkBPX oDMnA — Understanding is the new bottleneck — Geoffrey Litt, Notion
Transcript-backed resources
Transcript-backed resources
Transcript-backed resources
Transcript-backed resources
- youtube 1IdzkRVmWAA — How we taught agents to use good retrieval - Hanna Lichtenberg, Mixedbread AI
- youtube UM6sFg_jdlE — RAG is dead, right?? — Kuba Rogut, Turbopuffer
- youtube Akm1sqvWG4A — Bypassing the Multimodal Tax: Hybrid RAG, SQL RRF & UI Telemetry - Abed Matini, Ogilvy
- youtube zKk7sDMGDEQ — Benchmarking semantic code retrieval on Claude Code — Kuba Rogut, Turbopuffer
- youtube T0HhO4YtTfE — AI System Design: From Idea to Production - Apoorva Joshi, MongoDB
- youtube OXMMN XbxwA — Research to Reality: Bringing Frontier ML Research to Production - Vaidas Razgaitis, Higharc
- youtube wFTVEDYVJT0 — Building Agents with Amazon Nova Act and MCP - Du'An Lightfoot, Amazon (Full Workshop)
- youtube x5GEVnkuRw — Structuring the Unstructured - Cedric Clyburn, Red Hat
- youtube vh2VGuQ3zhY — The 100-Tool Agent Is a Trap - Sohail Shaikh & Ankush Rastogi, Prosodica
- youtube Jx4ZFEAq6bY — User Signal Dies at the Retrieval Boundary - Sonam Pankaj, StarlightSearch
- youtube dRmWYHuIJxM — We Cut 94% of AI Coding Tokens With a Local Code Index - Rajkumar Sakthivel, Tesco
- youtube XovaGv4f39A — When All Context Matters: Extended Cache Augmented Generation - Luis Romero-Sevilla, Orbis
- youtube btxGmN8RvNU — Your Agent's Biggest Lie: "I Searched the Web" — Rafael Levi, Bright Data
- youtube CDqzWpwkSls — Build AI Systems for Discernment, Not Approval - Angel Ortmann Lee, Duolingo
- youtube iNkFlCiij0U — The Art & Science of Benchmarking Agents — Vincent Chen, Snorkel AI
- youtube MpZzWMdmQCE — Your coding agent doesn't always follow your rules — Talha Sheikh, Checkout.com
- youtube EcqMYoIV57A — Why More Context Makes Your Agent Dumber and What to Do About It — Nupur Sharma, Qodo
Quote signals
- “And what I Turbo puffer what we think this actually means, you know if you break down rag into retrieval augmented generation, you know retrieval is not just vector search.” — youtube UM6sFg_jdlE
- “Um So what we're finding now is that a lot of people are no longer doing the simple rag you know the the Twitter quote unquote rag of just doing vector search once and throwing it into the context windows.” — youtube UM6sFg_jdlE
- “So you know not not a public benchmark but you can trust the numbers they give us.” — youtube UM6sFg_jdlE
Transcript-backed resources
Quote signals
This evidence graph consolidates scheduled talks, linked videos, transcripts, and slide-derived material connected to this topic.
Linked Sessions
- Vector Isn't Enough: Hybrid Search & Retrieval for AI Engineers
- The unreasonable effectiveness of BM25 for agentic search
- The Search Engine for the Agentic Web
- Where RL Will Take Search
- Autoresearch for Dense Retrieval: Test-Time Compute with Frozen Embedding Models
- « the era of (auto) research »
- Closing the Loop: An Autonomous AI Research Agent
- Teaching Agents to Search: Building Synthetic Training Pipelines with NVIDIA Data Designer
- Autoresearch in a Multi-Agent AI Village
- CrabRAG: Why Automated Assistants Need Graph Memory, Not More Tokens
Media Signals
youtube-xnXqpUW_Kp8— 5 slide-derived text signals- Slide-derived themes for
youtube-xnXqpUW_Kp8: humans, built, information, traditional, search, engines, type, simple. - Evidence links for
youtube-xnXqpUW_Kp8: youtube xnXqpUW_Kp8, youtube xnXqpUW_Kp8 slides, youtube xnXqpUW_Kp8 dense slides, youtube xnXqpUW_Kp8 reconstructed slides youtube-4sX_He5c4sI— 82,600 transcript words; 8 slide-derived text signals- Transcript signals for
youtube-4sX_He5c4sI: model, code, models, research, system, well, first, better. - Slide-derived themes for
youtube-4sX_He5c4sI: system, prompt, examples, tools, lots, claude, gets, smarter. - Evidence links for
youtube-4sX_He5c4sI: youtube 4sX_He5c4sI, youtube 4sX_He5c4sI transcript, youtube 4sX_He5c4sI slides, youtube 4sX_He5c4sI dense slides, youtube 4sX_He5c4sI reconstructed slides youtube-eW_vxrjvERk— 3 slide-derived text signals- Slide-derived themes for
youtube-eW_vxrjvERk: enter, conversations, github, memory, podcast, press, send, shit. - Evidence links for
youtube-eW_vxrjvERk: youtube eW_vxrjvERk, youtube eW_vxrjvERk slides, youtube eW_vxrjvERk dense slides, youtube eW_vxrjvERk reconstructed slides
Source Coverage
This table summarizes the local evidence already linked from this topic. It is a navigation aid, not a claim that every linked page has been fully reviewed.
| Evidence type | Count | Review note |
| --- | ---: | --- |
| other | 50 | Related pages outside the main evidence categories. |
| resources | 21 | Video/resource pages; check source status before treating as primary event evidence. |
| slides | 16 | OCR or reconstructed slide evidence; mark claims as OCR-derived unless image-reviewed. |
| talks | 24 | Official schedule pages; use for titles, speakers, tracks, and stated talk framing. |
| tools | 4 | Derived inventory pages; use as entity context, not independent proof. |
| transcripts | 1 | Transcript markdown; check session matching and caption quality. |
Talks
- 2026 07 01 session vector isn t enough hybrid search and retrieval for ai engineers
- 2026 06 29 jo kristian bergum the unreasonable effectiveness of bm25 for agentic search
- 2026 06 29 will bryk the search engine for the agentic web
- 2026 06 29 maximilian david rumpf where rl will take search
- 2026 06 30 han xiao autoresearch for dense retrieval test time compute with frozen embedding models
- 2026 06 30 elie bakouch the era of auto research
Resources
- youtube 1IdzkRVmWAA
- youtube UM6sFg_jdlE
- youtube Akm1sqvWG4A
- youtube zKk7sDMGDEQ
- youtube T0HhO4YtTfE
- youtube OXMMN XbxwA
Slides
- youtube HsxQICTLF84 slides
- youtube IQkVMvXQKLY slides
- youtube 1IdzkRVmWAA slides
- youtube 2e9ANoOEn28 slides
- youtube CLttOU7n6sI slides
- youtube UcYoMg 8 L8 slides
Transcripts
Tools
This table summarizes the local evidence already linked from this topic. It is a navigation aid, not a claim that every linked page has been fully reviewed.
Talks
Resources
Slides
Transcripts
Tools
This table summarizes the local evidence already linked from this topic. It is a navigation aid, not a claim that every linked page has been fully reviewed.
| resources | 22 | Video/resource pages; check source status before treating as primary event evidence. |
| tools | 5 | Derived inventory pages; use as entity context, not independent proof. |
Talks
Resources
Slides
Transcripts
Tools
This table summarizes the local evidence already linked from this topic. It is a navigation aid, not a claim that every linked page has been fully reviewed.
Talks
Resources
Slides
Transcripts
Tools
This table summarizes the local evidence already linked from this topic. It is a navigation aid, not a claim that every linked page has been fully reviewed.
Talks
Resources
Slides
Transcripts
Tools
This table summarizes the local evidence already linked from this topic. It is a navigation aid, not a claim that every linked page has been fully reviewed.
Talks
Resources
Slides
Transcripts
Tools
This table summarizes the local evidence already linked from this topic. It is a navigation aid, not a claim that every linked page has been fully reviewed.
| resources | 23 | Video/resource pages; check source status before treating as primary event evidence. |
Talks
Resources
Slides
Transcripts
Tools
This table summarizes the local evidence already linked from this topic. It is a navigation aid, not a claim that every linked page has been fully reviewed.
| resources | 24 | Video/resource pages; check source status before treating as primary event evidence. |
Talks
Resources
Slides
Transcripts
Tools
This table summarizes the local evidence already linked from this topic. It is a navigation aid, not a claim that every linked page has been fully reviewed.
Talks
Resources
Slides
Transcripts
Tools
Active Use Cases
- Research agents that cite and compare sources.
- Hybrid RAG over documents, SQL, UI telemetry, and web data.
- Semantic code retrieval for coding agents.
- Enterprise knowledge agents with source-grounded answers.
Slide-Derived Supporting Decks
- youtube CnA2lGfymY slides — "I've never seen anything scarier than an LLM with tool calls." — Erik Meijer aka @HeadinTheBox (32 extracted slide frames)
- youtube n97BCfyFIvw slides — "The engineer of the future is the person who is able to choose what is worth doing." — Addy Osmani (32 extracted slide frames)
These decks are slide/OCR support only; keep the article synopsis, origin, use cases, and schedule sections as the primary topic narrative.
Slide-Derived Scheduled Session Signals
- 2026 06 29 erik meijer in code they act in proof we trust — In Code They Act, In Proof We Trust
- 2026 06 30 addy osmani closing keynote — Closing Keynote