An agent is only as good as what it remembers, and most forget your decisions the moment the session ends. Teams paste the same context into every chat, or pile everything into one file the model half reads. Live, Param builds a second brain with Claude Code over notes, docs and Linear: it files what you capture, links it and answers with sources.
Why this matters
An agent is only as useful as what it remembers, and most forget your decisions the moment a session ends. Teams paste the same background into every chat or keep one huge file the model only half reads. A second brain gives agents and people one place where notes, docs and tickets are filed, linked and searchable, so answers come with sources you can check.
What happens in the hour
The problem4 min
Why pasting the same context into every chat stops working once a team relies on agents.
Built live40 min
Claude Code filing, tagging and linking notes, docs and issues; answers that cite their source; the brain served over MCP to other agents.
Questions16 min
What to put in your own team's brain first, and what to leave out.
What you will be able to do
Capture into one brain
Claude Code files, tags and links what you add as you work.
Answer with sources
Every answer cites the note or issue it came from.
Share the memory
Serve the brain over MCP so any agent can read from it.
The ideas behind it
Capture
Notes, documents and tickets go into one store with where they came from, without anyone tidying them first.
Linking
Related items are connected, so a question about a decision finds the ticket, the doc and the meeting note behind it.
Answers with sources
The agent answers from what it found and shows where, so a person can check the claim in one click.
Shared memory
The same brain serves the whole team, so one engineer's context helps the next one.
Where it fits
The parts of a production agent system, in the order the buildcamp builds them. The lit tiles are the ones this live lab builds; the Agentic AI Buildcamp for Engineers builds all of them.
Week 1Architect
Roles and modelsOne job per agent, a model chosen for that job, and a token budget.
OrchestrationA queue agents pick work from, with hand-offs a person can follow.
Context and memoryRetrieval with sources, memory across sessions, prompts laid out for the cache.
Tool callingTyped tools that act in GitHub, Linear and Google.
Week 2Build
MCP serversEvery tool behind an MCP gateway, scoped to the role that needs it.
Durable executionRuns that resume after a crash and never repeat a write.
Human in the loopA person approves anything that cannot be undone.
Week 3Secure and deploy
Identity and permissionsEach agent signs in as itself and acts on behalf of a user.
LLM gatewayRouting, caching and a budget on every model call.
Traces, evals and costEvery run traced, scored and charged to the agent that made it.
Multi-tenancyEach team or customer kept apart, in data and in the bill.
Week 4Govern and extend
Policy as codeRules checked on every action, not written in a document.
Signed skillsNew roles built from reviewed, signed skills.
Before you come
For
Engineers who repeat the same context to their agents every day, and tech leads who want decisions findable by people and agents.
You need
None. Code is shown, not required.
Track
Agentic AI for engineers who build and run it.
Questions
Is this the same as RAG?
Retrieval is one part of it. A second brain also decides what to keep, how items link and how memory is shared across people and sessions.
Where does the data live?
In files and services you control. The lab runs locally and shows what changes when a team shares it.
Does it work with tools other than Linear?
Yes. Linear is the example; any source with an API or an MCP server can feed the same brain.
Do I need to write code?
No. The build is shown live. Engineers can repeat it with Claude Code afterwards.
Is the Live Lab really free?
Yes. Live Labs are free on Maven. You sign up with your email and get the join link and the recording.
What if I cannot make it live?
Sign up anyway. Everyone who signs up gets the recording, so you can watch the build later and reply with questions.
Do I need to code along?
No. Most people watch the build and ask questions. Every step is shown, so you can repeat it on your own afterwards.