---
schema: 1
type: video
title: >-
  Everything You Need To Know About Agent Observability — Danny Gollapalli &
  Zubin Koticha, Raindrop
description: >-
  Agent failures do not look like normal software failures. In this workshop,
  the Raindrop team breaks down what it actually takes to monitor production
  agents, from explicit signals like tool errors, latency, and cost to fuzzier
  signals like user frustration, refusals, task failure, and capability gaps.
  The session covers how to move beyond evals toward real production
  observability, how to use classifiers, regex, and experiments to catch
  regressions, and how to instrument self-diagnostics so agents can report their
  own failures and strange behavior. If you're running agents in production,
  this is a practical framework for understanding what is going wrong and how to
  catch it early. Speaker info: - https://x.com/benhylak -
  https://www.linkedin.com/in/zkoticha -
  https://www.linkedin.com/in/joseph-daniel-gollapalli-a371a4138/ Timestamps
  0:14 Introduction and the problem of agent failures 1:48 Moving from evals to
  production monitoring 3:33 The two types of signals: explicit and implicit
  4:47 Using classifier signals for observability 6:38 Leveraging regex for
  signal detection 7:30 Using experiments to validate improvements 9:42 Q&A
  session: Statistical relevance and experimental design 16:07 Introduction to
  self-diagnostics 20:15 Workshop: Coding agent demonstration 24:01 Live demo:
  Triggering and handling tool failure 30:26 Best practices for self-diagnostic
  implementation 32:20 Q&A: Real-world use cases and triage 40:02 Q&A: Managing
  fast-paced experimentation 44:21 Q&A: Trace visualization and data export ---
  Watch along with the corrected transcript, summary, timestamps and resources
  on this talk's official AIE page:
  https://ai.engineer/talks/-aM2EDTiaMs-everything-you-need-to-know-about-agent-observability
  Join our upcoming AI Engineer conferences (see https://ai.engineer landing
  page): - AIEi Shanghai, November 5–6 — the first for meeting top Chinese labs
  and engineers - AIE CODE in SF, November 10–12 — the top AI Coding conference
  returns - AIEi Sydney, December 7–8 — the first AIEi run alongside NeurIPS
  2026 in Sydney ---
canonical: 'https://liquidgold.cc/watch/-aM2EDTiaMs'
markdown: 'https://liquidgold.cc/watch/-aM2EDTiaMs.md'
llms: 'https://liquidgold.cc/watch/-aM2EDTiaMs/llms.txt'
parent: 'https://liquidgold.cc/channel/aidotengineer.md'
channel: aidotengineer
updated: '2026-10-24T14:12:15.266Z'
language: null
children: []
videoId: '-aM2EDTiaMs'
durationSeconds: 3025
publishedAt: '2026-05-07T13:00:06.000Z'
searchChunks:
  - ordinal: 0
    text: null
    startSeconds: 26
    endSeconds: 108
    watch: 'https://liquidgold.cc/watch/-aM2EDTiaMs?t=26'
    heldReason: rights
  - ordinal: 1
    text: null
    startSeconds: 26
    endSeconds: 108
    watch: 'https://liquidgold.cc/watch/-aM2EDTiaMs?t=26'
    heldReason: rights
  - ordinal: 2
    text: null
    startSeconds: 119
    endSeconds: 252
    watch: 'https://liquidgold.cc/watch/-aM2EDTiaMs?t=119'
    heldReason: rights
  - ordinal: 3
    text: null
    startSeconds: 119
    endSeconds: 252
    watch: 'https://liquidgold.cc/watch/-aM2EDTiaMs?t=119'
    heldReason: rights
  - ordinal: 4
    text: null
    startSeconds: 119
    endSeconds: 252
    watch: 'https://liquidgold.cc/watch/-aM2EDTiaMs?t=119'
    heldReason: rights
  - ordinal: 5
    text: null
    startSeconds: 252
    endSeconds: 262
    watch: 'https://liquidgold.cc/watch/-aM2EDTiaMs?t=252'
    heldReason: rights
  - ordinal: 6
    text: null
    startSeconds: 262
    endSeconds: 352
    watch: 'https://liquidgold.cc/watch/-aM2EDTiaMs?t=262'
    heldReason: rights
  - ordinal: 7
    text: null
    startSeconds: 363
    endSeconds: 489
    watch: 'https://liquidgold.cc/watch/-aM2EDTiaMs?t=363'
    heldReason: rights
  - ordinal: 8
    text: null
    startSeconds: 363
    endSeconds: 489
    watch: 'https://liquidgold.cc/watch/-aM2EDTiaMs?t=363'
    heldReason: rights
  - ordinal: 9
    text: null
    startSeconds: 499
    endSeconds: 629
    watch: 'https://liquidgold.cc/watch/-aM2EDTiaMs?t=499'
    heldReason: rights
  - ordinal: 10
    text: null
    startSeconds: 499
    endSeconds: 629
    watch: 'https://liquidgold.cc/watch/-aM2EDTiaMs?t=499'
    heldReason: rights
  - ordinal: 11
    text: null
    startSeconds: 640
    endSeconds: 763
    watch: 'https://liquidgold.cc/watch/-aM2EDTiaMs?t=640'
    heldReason: rights
  - ordinal: 12
    text: null
    startSeconds: 763
    endSeconds: 774
    watch: 'https://liquidgold.cc/watch/-aM2EDTiaMs?t=763'
    heldReason: rights
  - ordinal: 13
    text: null
    startSeconds: 774
    endSeconds: 863
    watch: 'https://liquidgold.cc/watch/-aM2EDTiaMs?t=774'
    heldReason: rights
  - ordinal: 14
    text: null
    startSeconds: 982
    endSeconds: 1091
    watch: 'https://liquidgold.cc/watch/-aM2EDTiaMs?t=982'
    heldReason: rights
  - ordinal: 15
    text: null
    startSeconds: 982
    endSeconds: 1091
    watch: 'https://liquidgold.cc/watch/-aM2EDTiaMs?t=982'
    heldReason: rights
  - ordinal: 16
    text: null
    startSeconds: 1091
    endSeconds: 1102
    watch: 'https://liquidgold.cc/watch/-aM2EDTiaMs?t=1091'
    heldReason: rights
  - ordinal: 17
    text: null
    startSeconds: 1102
    endSeconds: 1221
    watch: 'https://liquidgold.cc/watch/-aM2EDTiaMs?t=1102'
    heldReason: rights
  - ordinal: 18
    text: null
    startSeconds: 1323
    endSeconds: 1442
    watch: 'https://liquidgold.cc/watch/-aM2EDTiaMs?t=1323'
    heldReason: rights
  - ordinal: 19
    text: null
    startSeconds: 1613
    endSeconds: 1704
    watch: 'https://liquidgold.cc/watch/-aM2EDTiaMs?t=1613'
    heldReason: rights
  - ordinal: 20
    text: null
    startSeconds: 1715
    endSeconds: 1804
    watch: 'https://liquidgold.cc/watch/-aM2EDTiaMs?t=1715'
    heldReason: rights
  - ordinal: 21
    text: null
    startSeconds: 2014
    endSeconds: 2113
    watch: 'https://liquidgold.cc/watch/-aM2EDTiaMs?t=2014'
    heldReason: rights
  - ordinal: 22
    text: null
    startSeconds: 2014
    endSeconds: 2113
    watch: 'https://liquidgold.cc/watch/-aM2EDTiaMs?t=2014'
    heldReason: rights
  - ordinal: 23
    text: null
    startSeconds: 2234
    endSeconds: 2321
    watch: 'https://liquidgold.cc/watch/-aM2EDTiaMs?t=2234'
    heldReason: rights
  - ordinal: 24
    text: null
    startSeconds: 2331
    endSeconds: 2418
    watch: 'https://liquidgold.cc/watch/-aM2EDTiaMs?t=2331'
    heldReason: rights
  - ordinal: 25
    text: null
    startSeconds: 2517
    endSeconds: 2528
    watch: 'https://liquidgold.cc/watch/-aM2EDTiaMs?t=2517'
    heldReason: rights
  - ordinal: 26
    text: null
    startSeconds: 2630
    endSeconds: 2734
    watch: 'https://liquidgold.cc/watch/-aM2EDTiaMs?t=2630'
    heldReason: rights
  - ordinal: 27
    text: null
    startSeconds: 2744
    endSeconds: 2854
    watch: 'https://liquidgold.cc/watch/-aM2EDTiaMs?t=2744'
    heldReason: rights
---


# Everything You Need To Know About Agent Observability — Danny Gollapalli & Zubin Koticha, Raindrop

Agent failures do not look like normal software failures. In this workshop, the Raindrop team breaks down what it actually takes to monitor production agents, from explicit signals like tool errors, latency, and cost to fuzzier signals like user frustration, refusals, task failure, and capability gaps. The session covers how to move beyond evals toward real production observability, how to use classifiers, regex, and experiments to catch regressions, and how to instrument self-diagnostics so agents can report their own failures and strange behavior. If you're running agents in production, this is a practical framework for understanding what is going wrong and how to catch it early. Speaker info: - https://x.com/benhylak - https://www.linkedin.com/in/zkoticha - https://www.linkedin.com/in/joseph-daniel-gollapalli-a371a4138/ Timestamps 0:14 Introduction and the problem of agent failures 1:48 Moving from evals to production monitoring 3:33 The two types of signals: explicit and implicit 4:47 Using classifier signals for observability 6:38 Leveraging regex for signal detection 7:30 Using experiments to validate improvements 9:42 Q\&A session: Statistical relevance and experimental design 16:07 Introduction to self-diagnostics 20:15 Workshop: Coding agent demonstration 24:01 Live demo: Triggering and handling tool failure 30:26 Best practices for self-diagnostic implementation 32:20 Q\&A: Real-world use cases and triage 40:02 Q\&A: Managing fast-paced experimentation 44:21 Q\&A: Trace visualization and data export --- Watch along with the corrected transcript, summary, timestamps and resources on this talk's official AIE page: https://ai.engineer/talks/-aM2EDTiaMs-everything-you-need-to-know-about-agent-observability Join our upcoming AI Engineer conferences (see https://ai.engineer landing page): - AIEi Shanghai, November 5–6 — the first for meeting top Chinese labs and engineers - AIE CODE in SF, November 10–12 — the top AI Coding conference returns - AIEi Sydney, December 7–8 — the first AIEi run alongside NeurIPS 2026 in Sydney ---

Canonical: https://liquidgold.cc/watch/-aM2EDTiaMs · 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.
