proof-after-planning layer
AIE Build Scout
A source-backed field OS for turning AI Engineer schedule pressure, hallway objections, and demo opportunities into a build receipt you can ship before the next room.
official, WFSF, AIEWF, schedule skill
551 speakers, 68 side events, counts reconciled
agents, context, authority, realtime, interface
Competition
Official schedule-app challenge
Build your own schedule app with official AIE data and share it for a chance to be featured.
The official schedule page links sessions, speakers, MCP, iCal, and embeddings data for app builders.
A public schedule app that turns the 568-session JSON feed, 573 rendered schedule entries, 551 speakers, 68 side events, and personal goals into an inspectable plan.
Itinerary packet
Primary, backup, skip, and follow-up in one view.
Start with the room or opportunity that changes the build.
Keep one fallback if the room is crowded or the plan conflicts.
Skip anything that cannot change the proof gate.
Share only after the source, route, and receipt are inspectable.
Generated plan
Make the build inspectable before it is impressive.
Mission control
Turn the official schedule firehose into one executable mission.
The live schedule is too large to browse honestly in the hallway. Pick the pressure cluster, then leave with the room to attend, the skeptical ask, the linked Chopshopr route, and the proof artifact that must exist before follow-up.
Agent reliability mission
Official sessions JSON v4859 shows 366 agent matches and 132 eval/evaluation matches in a full-record keyword scan.
- Operator move
- Treat one agent session as the primary room, one eval or harness session as backup, and one skeptical hallway ask as the test name.
- Proof artifact
- An agent receipt with input, tool boundary, blocked action, eval result, retry path, and follow-up question.
- Hallway ask
- What failure would make your current agent eval suite untrustworthy?
- Ship gate
- A failing scenario becomes a named test, the route exposes the test result, and the export packet includes the recovery path.
Attendee pain map
Do not build another schedule clone. Build the receipt after the schedule.
Current public data is already noisy: the official schedule UI, sessions JSON, and community apps do not all present the same count. The useful attendee layer is the part that says what to skip, what to attend, what to ask, and what public proof changed because of it.
Build receipt loop
Pick the pressure in the room, then leave with a proof packet.
This is the differentiator: the lab turns AIE attendance into a same-day loop of room, objection, proof gate, route change, and follow-up. Planning is only the first move.
Agent production loop
Everyone is talking agents; few can show retries, evals, memory, tool boundaries, and operator handoff in one proof.
- Source signal
- Official schedule and side events cluster around Agentic Engineering, Harness Engineering, Software Factories, and agents in production.
- Next action
- Pick one agent reliability session, one harness/evals backup, and one side event where production operators will be skeptical.
- Proof artifact
- A route or demo packet that shows input, tool call, blocked action, retry/eval result, and the follow-up question as a named gate.
- Ask in the room
- What would make you trust this agent after the demo fails once?
Agent production loop: route proof plus follow-up gate.
Featured submission teardown
Steal the parts that make people actually use the app.
The official page now features community schedule apps. The useful lesson is not their styling; it is the loop: current data, personal choices, backups, conflicts, social follow-through, and an obvious path from browsing to proof.
Cooked field plan
Pick a goal, then make the conference behave like a product loop.
This is the piece the old Scout missed: a schedule app is not only a calendar. It is a decision engine for what to attend, what to skip, who to meet, and what proof to ship before the next room.
Agent shipping day plan
Builders trying to leave with a public artifact.
Expert defect audit
Different experts find different product defects.
The product gets better when each lens forces a specific correction. Select a lens to see the defect and the iteration it demands.
Ambition, speed, and compounding distribution
Defect: The old Scout proved source discipline but did not compound into attendee utility or distribution.
Iteration: Add a current receipt, borrow featured-app planning primitives, and make every export usable as a share loop.
Challenge action map
Action the AIE field through Chopshopr surfaces.
Each card turns an AI Engineer build challenge into a public route, lab, or article with a proof gate. The goal is a working artifact, not an event recap.
Frontier repo radar
Turn popular WebGL and WebGPU repos into build moves.
The new radar ranks public Three.js, WebGL, WebGPU, canvas, game, data, and AI-native repos, then forces each winner through a Chopshopr move, an AI Engineer Lab move, and a proof gate.
Open Frontier WebGL Radar