Autoresearch
Synopsis
AutoResearch is the use of agents to search, read, compare, synthesize, benchmark, and sometimes design follow-up experiments over a body of evidence. In the WF2026 Autoresearch track, the concept spans automated AI research, dense retrieval with test-time compute over frozen embedding models, autonomous research-agent loops, reflective self-improvement of context and model weights, kernel optimization, and production pathways from frontier ML research into usable systems. The goal is not just summarization; it is repeatable research workflow support with source tracking, uncertainty management, evaluation, and clear next-step planning.
Origin And Context
It grew from literature search, systematic review methods, research assistants, web search, RAG, benchmarking, and scientific-discovery tooling. LLM agents added the ability to decompose questions, inspect sources, generate hypotheses, compare evidence, and produce structured research artifacts. The connected WF2026 material places AutoResearch in a broader shift from one-off retrieval toward closed-loop systems: agents that gather evidence, run or propose tests, improve their own harnesses, and move research ideas toward production workflows.
Why It Matters
Research work is expensive because it involves discovery, filtering, evidence comparison, synthesis under uncertainty, and judgment about what to try next. The connected sessions make the topic concrete: Richard Socher frames automated AI research as an emerging research direction, Han Xiao ties autoresearch to retrieval quality and test-time compute, Tim Sweeney focuses on autonomous research-agent loops, and Lakshya Agrawal connects self-improvement to context, harnesses, and model weights. Agents can accelerate the mechanical parts, but only if they preserve citations, distinguish claims from evidence, and expose gaps instead of hiding uncertainty behind polished prose.
How To Use It
Start with a clear research question, source-specific retrieval, and an explicit record of search terms, inclusion criteria, and excluded evidence. Keep a claim-evidence table that separates official schedule facts, transcript-backed observations, slide/OCR-derived notes, interpretations, and open questions. Use agentic search and memory for multi-step exploration, but pair them with agent evaluations, benchmark design, and human review before treating outputs as conclusions. For engineering research, connect the synthesis to reproducible artifacts: experiments, eval harnesses, retrieval tests, kernel benchmarks, or implementation plans.
Where It Is Useful
AutoResearch is useful for technical due diligence, literature reviews, market maps, competitive analysis, financial-compliance document correlation, product discovery, and engineering design investigations. In this wiki, it is also a method for conference intelligence: the official Autoresearch livestream, extracted slides/OCR, scheduled talks, and transcript-backed resource pages can be compared to identify recurring claims, tools, research patterns, and unanswered questions across talks.
When To Use It
Use it when the answer depends on multiple sources, evolving evidence, or repeated comparison across papers, products, transcripts, benchmarks, or implementation patterns. It is especially relevant when a team needs a source-grounded briefing, a research map, or an experiment plan rather than a single answer. Avoid relying on it as a black-box oracle for high-stakes conclusions; the connected material repeatedly points toward closed-loop research systems, but those loops still need traceable evidence, evaluation, and human judgment.
Active Use Cases
- Evidence-grounded briefing docs and source maps.
- Research agents that compare papers, products, or implementation patterns.
- Experiment-planning support for AI and data teams.
- Conference or domain wiki synthesis from talks, transcripts, and slides.
Related Scheduled Sessions
- 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 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 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 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 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 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 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 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 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)
- 2026 06 30 ishan anand will ai predict people like we predict the weather alternate title a field guide to synthetic personas for market research — Will AI predict people like we predict the weather? (alternate title “A field guide to synthetic personas for market research”); Ishan Anand (Day 3 — Session Day 2 · 2:50pm-3:10pm · Computer Use; official schedule)
- 2026 06 30 deepak pathak frontier robotics research — Frontier Robotics Research; Deepak Pathak (Day 3 — Session Day 2 · 1:55pm-2:15pm · Robotics & World Models; official schedule)
- 2026 06 30 zhengyao jiang an ai agent became the 1 contributor in openai s hiring challenge — An AI Agent Became the #1 Contributor in OpenAI's Hiring Challenge; Zhengyao Jiang (Day 3 — Session Day 2 · 1:55pm-2:15pm · Autoresearch; official schedule)
- 2026 06 29 sonar expo welcome speech — Expo Welcome Speech; Sonar, Extend AI (Day 1 — Workshop Day · 6:00pm-6:15pm · Expo Stage 3; related YouTube resource; via youtube 4sX_He5c4sI)
- 2026 06 29 charlie guo cooking with codex — Cooking with Codex; Charlie Guo, Gabriel Chua (Day 1 — Workshop Day · 9:00am-11:00am · Workshops Day 1; related YouTube resource; via youtube dvft0Gp9sEE)
- 2026 06 29 charlie guo voice agents can just do things — Voice Agents Can Just Do Things; Charlie Guo (Day 2 — Session Day 1 · 11:40am-12:00pm · Voice & Realtime AI; related YouTube resource; via youtube dvft0Gp9sEE)
- 2026 06 29 doug guthrie advanced workshop mastering ai observability — Advanced workshop: Mastering AI Observability; Doug Guthrie (Day 1 — Workshop Day · 9:00am-11:00am · Track 9; related YouTube resource; via youtube bk0TmxoZlUY)
- 2026 06 30 thariq shihipar field guide to fable — Field Guide to Fable; Thariq Shihipar (Day 3 — Session Day 2 · 9:05am-9:25am · Autoresearch; official schedule)
- 2026 06 30 antje barth perception agents — Perception Agents; Antje Barth (Day 3 — Session Day 2 · 9:45am-10:05am · Autoresearch; official schedule)
- 2026 06 30 laurie voss evals track intro — Evals Track Intro; Laurie Voss, Aparna Dhinakaran (Day 3 — Session Day 2 · 10:25am-10:30am · Autoresearch; official schedule)
- 2026 06 30 lakshya agrawal self improvement of context harness and model weights through reflective optimization — Self-Improvement of Context, Harness, and Model Weights through Reflective Optimization; Lakshya Agrawal (Day 3 — Session Day 2 · 2:25pm-2:45pm · Autoresearch; official schedule)
Related People
- Laurie Voss
- Zhengyao Jiang
- Charlie Guo
- Tim Sweeney
- Dixing Xu
- Vayum Arora
- Dhruv Srikanth
- Elie Bakouch
- Erina Karati
- Arunachalam Manikandan
- Han Xiao
- Tejas Bhakta
- Roland Gavrilescu
- Julian Bright
- Brendan Rappazzo
- Benoit Schillings
- Richard Socher
- Stefania Druga
- Zubin Aysola
- Valeria Wu Fon
- Tom Ouyang
- Ishan Anand
- Deepak Pathak
- Sonar
Related Companies
- Weco AI
- Google DeepMind
- Arize AI
- Together AI
- OpenAI
- Weights & Biases by CoreWeave
- Introspection
- Artificial Analysis
- DoorDash
- Browserbase
- Superlinked
- Atlassian
- FriendliAI
- Coreweave
- Prime Intellect
- Supercell
- University of Minnesota
- Elastic
Transcript And Resource Support
Transcript-backed resources
- youtube OXMMN XbxwA — Research to Reality: Bringing Frontier ML Research to Production - Vaidas Razgaitis, Higharc
- youtube aHhB3sjGjkI — Agents Building Agents - Alfonso Graziano, Nearform
- youtube zMiSRliEzv4 — Self Driving Products: Product Signals to Pull Requests — Joshua Snyder, PostHog
- youtube iNkFlCiij0U — The Art & Science of Benchmarking Agents — Vincent Chen, Snorkel AI
- youtube fWXJM J0ZB8 — Frontier results, on device - RL Nabors, Arize
- youtube u rJwPPU3QA — How to talk to statues — Joe Reeve, ElevenLabs
- youtube IJXjTLPzvAU — The Miranda Hypothesis: How Hamilton Poisoned Persona Evals - Jacob E. Thomas, Results Gen
- youtube 4sX_He5c4sI — WF2026: Autoresearch & Keynotes ft. Anthropic, Google DeepMind, Amazon AGI, Sonar, Arena, Recursive
- youtube 0S8xe9ftGTM — 6 Things to Know about AIE World's Fair 2026
- youtube Iwe_RY fYgI — AI-Driven Multi-Document Correlation for Financial Compliance - Varsha Shah, Independent
- youtube akk6KRlcwW4 — OpenClaw in Your Hand: Building a Physical AI Terminal - Lech Kalinowski, Callstack
- youtube pSto5YaNGUo — The Agentic AI Engineer - Benedikt Sanftl, Mutagent
- youtube dvft0Gp9sEE — Analyzing 10,000 Sales Calls With AI In 2 Weeks — Charlie Guo
- youtube bk0TmxoZlUY — Evals 101 — Doug Guthrie, Braintrust
- youtube hqHC6Z_lXyo — 20 days of compute vs 7 hours: rethinking what state-of-the-art means — Bertrand Charpentier, Pruna
- youtube whue9_YquGA — Building an Autonomous Engineering Org - Angie Jones, Agentic AI Foundation
- youtube hVJOnuhFmTA — The Prompt Is Still a Punch Card - Ted Johnson, JoinIn AI
Livestream Source
- youtube 4sX_He5c4sI — official WF2026 Autoresearch and keynote livestream.
- youtube 4sX_He5c4sI slides — extracted slide/OCR deck for the livestream.