Launch Directory/AI & ML/MemCurrent — memory for production agents
MemCurrent — memory for production agents

MemCurrent — memory for production agents · Launch Video Breakdown: Hook, Pacing & Motion Design

0:00 · The Hook · The Amnesiac Agent Problem
0:00 / 0:00
AI & MLProductionOctober 20, 2026

Pain Point Demonstration

6.75

Cuts / min

1:20

Runtime

0K

Views

274

Likes

0

Reposts

0

Replies

Scene-by-scene timeline & spoken transcript

  1. The Hook

    The Amnesiac Agent Problem

    “Most agents start every conversation from zero. They forget what users told them and redo work they've already done.”

    On screen
    Support agent Conversation 1 Let's fix that bug users reported last week. search_issues("login fails") query logs.auth_errors --since 7d read auth/session.py run tests/test_oauth.py Conversation 2 Any progress on that login bug? Which bug? Let me look into it. AGAIN
    Camera
    Static isometric framing with subtle floating card hover.
    Motion
    UI card morphing from Conversation 1 to Conversation 2 with repeating execution badges.
  2. Product Reveal

    Introducing Agent Memory

    “That's why we built Agent Memory into Agent Bricks. It has three pieces, all built on Lakebase.”

    On screen
    AGENT MEMORY Your agent Any framework or harness LAKEBASE - IN YOUR WORKSPACE
    Camera
    Subtle zoom out on an isometric grid diagram.
    Motion
    Card expansion revealing architectural dashed outline slots.
  3. Feature Teaser

    Sessions & Managed Memory

    “Managed sessions record each conversation, turn by turn. Managed memory keeps what's worth knowing about each user: preferences, facts, and workflows.”

    On screen
    01 / SESSIONS append each turn Managed sessions Full conversation history session_a session_b session_c turn 1 turn 2 turn 3 02 / MEMORY Managed memory What's worth keeping PER USER /memories/ preferences/ facts/ workflows/ save recall
    Camera
    Smooth pan across the isometric architectural canvas.
    Motion
    Staggered sequential container pop-ins with animated connector lines.
  4. Feature Teaser

    Background Memory Pipelines

    “Memory pipelines connect the two. In the background, they distill sessions into memory, merge duplicates, and keep memory files easy to access.”

    On screen
    03 / PIPELINES Memory pipeline Runs in the background distill MEMORY PIPELINE Sessions in, durable memory out. SESSIONS Users can't log in since Tuesday. Logs show the token expires early. Login is still failing for users. The fix goes in auth/session.py MEMORY Login bug since Tuesday Cause: token expires early Fix: auth/session.py facts/login-bug.md
    Camera
    Detail zoom focusing on the distillation transformation panel.
    Motion
    Data node transformation sliding raw logs into structured markdown entries.
  5. Feature Teaser

    Empirical Benchmark Results

    “The Databricks research team found that agents with memory passed three times as many tasks, using thirty-one percent fewer tokens.”

    On screen
    DATABRICKS RESEARCH 3x more tasks passed No memory 22.3% With memory 68.4% 31% fewer tokens per task No memory 603K With memory 418K Databricks internal memory benchmark: 114 enterprise data tasks drawn from real agent sessions.
    Camera
    Static dark UI view with left-to-right bar animation.
    Motion
    Animated progress bar fill with spring-eased numeric count-up.
  6. Feature Teaser

    Simple CLI Configuration

    “Hooking it up is easy: bind a session store and a memory store, then create a memory pipeline between the stores. Now your memories are stored in Lakebase, and Databricks automatically synthesizes new ones in the background.”

    On screen
    AGENT BRICKS CLI $ agentbricks sessions bind support-agent-sessions $ agentbricks memory bind support-agent-memory $ agentbricks memory pipeline create --session-store support-agent-sessions --memory-store support-agent-memory --trigger scheduled support-agent Session store Memory store Memory pipeline scheduled SYNTHESIZING IN THE BACKGROUND
    Camera
    Split view transition between terminal console and live topology graph.
    Motion
    Typewriter terminal simulation triggering synchronized graph connection states.
  7. Product Reveal

    Instant Recall in Action

    “Your agent's next conversation starts with what it learned from the last one.”

    On screen
    Support agent Next conversation Let's finish that login bug fix. Recalled from memory: facts/login-bug.md On it. The fix goes in auth/session.py edit auth/session.py ONE CALL
    Camera
    Centered framing on the final active conversation bubble.
    Motion
    Pill tag notification expansion with single-shot tool execution trigger.
  8. Call to Action

    Availability & Brand Outro

    “Agent Memory, now in beta on Agent Bricks.”

    On screen
    Agent Memory Now in Beta on Agent Bricks databricks
    Camera
    Static title screen followed by rotating geometric Databricks logo lockup.
    Motion
    Clean title fade-in followed by branded tile shape rotation and snap.

Related AI & ML Product Launches

Explore all AI & ML launches →
Grok Bot 1.0 — AI teammates for your team
@bot
AI & ML
@bot
Grok Bot 1.0 — AI teammates for your team@bot
Tauvus
@tavus
AI & ML
@tavus
Tauvus@tavus
Fish Audio
@FishAudio
AI & ML
@FishAudio
Fish Audio@FishAudio