---
schema: 1
type: video
title: 'Wearing the Agent: From Group Chats to Glasses — Sai Krishna Rallabandi'
description: >-
  Almost every agent today is built for one user, and Sai Krishna Rallabandi has
  spent about eight months on what breaks when you drop one into a group chat
  instead. Running a personal agent across a real group, and eventually onto
  glasses, forces two hard problems. The first is memory: conversations that
  last for days across several people, where the agent has to track who said
  what and surface only what is relevant without bloating its context. The
  second is security, and it is where the talk digs in. He points to a finding
  that two skills, each safe on its own, can collide once they share
  infrastructure, so a reporting agent leaks information at low frequency. His
  defense is to have an agent read everything and design a guard, then fine tune
  a small model with a per user adapter that only releases information
  appropriate to the context and catches prompt injection that regex would miss.
  The through line is that group and wearable settings break the single user
  assumptions most agent harnesses are built on, so memory and security have to
  be redesigned around the group. Speaker info: - https://x.com/Saikallis9012 -
  https://www.linkedin.com/in/sai-krishna-rallabandi-8595418b/ -
  https://saikrishnarallabandi.github.io/ Timestamps: 0:00 - Introduction:
  agents for groups, not one user 1:28 - Why single user agents fall short 2:45
  - Eight months in a real group chat 3:47 - What changes when the agent joins a
  group 5:17 - Two problems: guarding and memory 5:55 - Securing an agentic
  system 7:44 - When two safe skills collide 9:27 - Designing a guard agent
  10:21 - A small model with per user adapters 11:49 - Catching prompt injection
  12:29 - Designing memory for groups 15:29 - Context growth and token cost
canonical: >-
  https://liquidgold.cc/channel/aidotengineer/explore/video/wearing-the-agent-from-group-chats-to-glasses-sai-k-27868c92ab1b
markdown: >-
  https://liquidgold.cc/channel/aidotengineer/video/wearing-the-agent-from-group-chats-to-glasses-sai-k-27868c92ab1b.md
llms: null
parent: 'https://liquidgold.cc/channel/aidotengineer.md'
channel: aidotengineer
updated: '2026-10-24T06:29:43.230Z'
language: null
children: []
videoId: s67bE2Ur3bY
durationSeconds: 1149
publishedAt: '2026-07-29T22:58:19.000Z'
searchChunks:
  - ordinal: 0
    text: null
    startSeconds: 25
    endSeconds: 91
    watch: 'https://liquidgold.cc/watch/s67bE2Ur3bY?t=25'
    heldReason: rights
  - ordinal: 1
    text: null
    startSeconds: 91
    endSeconds: 102
    watch: 'https://liquidgold.cc/watch/s67bE2Ur3bY?t=91'
    heldReason: rights
  - ordinal: 2
    text: null
    startSeconds: 102
    endSeconds: 172
    watch: 'https://liquidgold.cc/watch/s67bE2Ur3bY?t=102'
    heldReason: rights
  - ordinal: 3
    text: null
    startSeconds: 183
    endSeconds: 227
    watch: 'https://liquidgold.cc/watch/s67bE2Ur3bY?t=183'
    heldReason: rights
  - ordinal: 4
    text: null
    startSeconds: 183
    endSeconds: 227
    watch: 'https://liquidgold.cc/watch/s67bE2Ur3bY?t=183'
    heldReason: rights
  - ordinal: 5
    text: null
    startSeconds: 240
    endSeconds: 317
    watch: 'https://liquidgold.cc/watch/s67bE2Ur3bY?t=240'
    heldReason: rights
  - ordinal: 6
    text: null
    startSeconds: 240
    endSeconds: 317
    watch: 'https://liquidgold.cc/watch/s67bE2Ur3bY?t=240'
    heldReason: rights
  - ordinal: 7
    text: null
    startSeconds: 373
    endSeconds: 464
    watch: 'https://liquidgold.cc/watch/s67bE2Ur3bY?t=373'
    heldReason: rights
  - ordinal: 8
    text: null
    startSeconds: 477
    endSeconds: 570
    watch: 'https://liquidgold.cc/watch/s67bE2Ur3bY?t=477'
    heldReason: rights
  - ordinal: 9
    text: null
    startSeconds: 477
    endSeconds: 570
    watch: 'https://liquidgold.cc/watch/s67bE2Ur3bY?t=477'
    heldReason: rights
  - ordinal: 10
    text: null
    startSeconds: 570
    endSeconds: 580
    watch: 'https://liquidgold.cc/watch/s67bE2Ur3bY?t=570'
    heldReason: rights
  - ordinal: 11
    text: null
    startSeconds: 580
    endSeconds: 624
    watch: 'https://liquidgold.cc/watch/s67bE2Ur3bY?t=580'
    heldReason: rights
  - ordinal: 12
    text: null
    startSeconds: 624
    endSeconds: 634
    watch: 'https://liquidgold.cc/watch/s67bE2Ur3bY?t=624'
    heldReason: rights
  - ordinal: 13
    text: null
    startSeconds: 634
    endSeconds: 714
    watch: 'https://liquidgold.cc/watch/s67bE2Ur3bY?t=634'
    heldReason: rights
  - ordinal: 14
    text: null
    startSeconds: 634
    endSeconds: 714
    watch: 'https://liquidgold.cc/watch/s67bE2Ur3bY?t=634'
    heldReason: rights
  - ordinal: 15
    text: null
    startSeconds: 749
    endSeconds: 759
    watch: 'https://liquidgold.cc/watch/s67bE2Ur3bY?t=749'
    heldReason: rights
  - ordinal: 16
    text: null
    startSeconds: 759
    endSeconds: 938
    watch: 'https://liquidgold.cc/watch/s67bE2Ur3bY?t=759'
    heldReason: rights
  - ordinal: 17
    text: null
    startSeconds: 759
    endSeconds: 938
    watch: 'https://liquidgold.cc/watch/s67bE2Ur3bY?t=759'
    heldReason: rights
  - ordinal: 18
    text: null
    startSeconds: 938
    endSeconds: 950
    watch: 'https://liquidgold.cc/watch/s67bE2Ur3bY?t=938'
    heldReason: rights
  - ordinal: 19
    text: null
    startSeconds: 950
    endSeconds: 1149
    watch: 'https://liquidgold.cc/watch/s67bE2Ur3bY?t=950'
    heldReason: rights
  - ordinal: 20
    text: null
    startSeconds: 950
    endSeconds: 1149
    watch: 'https://liquidgold.cc/watch/s67bE2Ur3bY?t=950'
    heldReason: rights
  - ordinal: 21
    text: null
    startSeconds: 950
    endSeconds: 1149
    watch: 'https://liquidgold.cc/watch/s67bE2Ur3bY?t=950'
    heldReason: rights
---


# Wearing the Agent: From Group Chats to Glasses — Sai Krishna Rallabandi

Almost every agent today is built for one user, and Sai Krishna Rallabandi has spent about eight months on what breaks when you drop one into a group chat instead. Running a personal agent across a real group, and eventually onto glasses, forces two hard problems. The first is memory: conversations that last for days across several people, where the agent has to track who said what and surface only what is relevant without bloating its context. The second is security, and it is where the talk digs in. He points to a finding that two skills, each safe on its own, can collide once they share infrastructure, so a reporting agent leaks information at low frequency. His defense is to have an agent read everything and design a guard, then fine tune a small model with a per user adapter that only releases information appropriate to the context and catches prompt injection that regex would miss. The through line is that group and wearable settings break the single user assumptions most agent harnesses are built on, so memory and security have to be redesigned around the group. Speaker info: - https://x.com/Saikallis9012 - https://www.linkedin.com/in/sai-krishna-rallabandi-8595418b/ - https://saikrishnarallabandi.github.io/ Timestamps: 0:00 - Introduction: agents for groups, not one user 1:28 - Why single user agents fall short 2:45 - Eight months in a real group chat 3:47 - What changes when the agent joins a group 5:17 - Two problems: guarding and memory 5:55 - Securing an agentic system 7:44 - When two safe skills collide 9:27 - Designing a guard agent 10:21 - A small model with per user adapters 11:49 - Catching prompt injection 12:29 - Designing memory for groups 15:29 - Context growth and token cost

Canonical: https://liquidgold.cc/channel/aidotengineer/explore/video/wearing-the-agent-from-group-chats-to-glasses-sai-k-27868c92ab1b · Channel: [AI Engineer](https://liquidgold.cc/channel/aidotengineer.md)

> Search chunk text is held until public-source provenance is verified.

## Document chunks

> Missing: no stored document chunks are available.