Upgrade to Pro — share decks privately, control downloads, hide ads and more …

IVRyエンジニア忘年LT大会2024 LLM監視の最前線

Sponsored · Your Podcast. Everywhere. Effortlessly. Share. Educate. Inspire. Entertain. You do you. We'll handle the rest. →
Avatar for Hiroyuki Moriya Hiroyuki Moriya
December 11, 2024
1
400

IVRyエンジニア忘年LT大会2024 LLM監視の最前線

Avatar for Hiroyuki Moriya

Hiroyuki Moriya

December 11, 2024
Tweet

More Decks by Hiroyuki Moriya

See All by Hiroyuki Moriya
少人数プロダクト開発において SRE業務を増やさない工夫
Avatar for Hiroyuki Moriya gekko0114
0
610
音声データ解析パイプラインの Software Engineering / Context Engineering
Avatar for Hiroyuki Moriya gekko0114
0
270
LLM Observabilityによる 対話型音声AIアプリケーションの安定運用
Avatar for Hiroyuki Moriya gekko0114
2
480
kueueに新しいPriorityClassを足した話
Avatar for Hiroyuki Moriya gekko0114
0
780
JobSet超入門
Avatar for Hiroyuki Moriya gekko0114
1
1.1k

Featured

See All Featured
Java REST API Framework Comparison - PWX 2021
Avatar for Matt Raible mraible
34
9.2k
Leadership Guide Workshop - DevTernity 2021
Avatar for David Neal reverentgeek
1
250
The untapped power of vector embeddings
Avatar for Frank van Dijk frankvandijk
2
1.6k
Abbi's Birthday
Avatar for Violet Bock coloredviolet
2
5.6k
YesSQL, Process and Tooling at Scale
Avatar for Rocio Delgado rocio
174
15k
Agile Leadership in an Agile Organization
Avatar for Dr. Kim W Petersen kimpetersen
PRO
0
120
Put a Button on it: Removing Barriers to Going Fast.
Avatar for kastner kastner
60
4.2k
HU Berlin: Industrial-Strength Natural Language Processing with spaCy and Prodigy
Avatar for Ines Montani inesmontani
PRO
0
280
The Anti-SEO Checklist Checklist. Pubcon Cyber Week
Avatar for Ryan Jones ryanjones
0
100
Chasing Engaging Ingredients in Design
Avatar for Sebastian Deterding codingconduct
0
150
Balancing Empowerment & Direction
Avatar for Lara Hogan lara
5
960
VelocityConf: Rendering Performance Case Studies
Avatar for Addy Osmani addyosmani
333
24k

Transcript

  1. confidencial LLM؂ࢹͷ࠷લઢ IVRy ΤϯδχΞ๨೥LTେձ 2024/12/11 Moriya Hiroyuki

  2. confidencial 2 Զͷ໊લ͸ɺߴߍੜ୳ఁ޻౻৽Ұ

  3. confidencial 3 Զͷ໊લ͸ɺߴߍੜ୳ఁ޻౻৽Ұ AIΤϯδχΞͱͯ͠ಇ͖࢝ΊͨԶ͸ɺLLMΛ࢖ͬͯɺͨ͘͞Μ͓ۚΛՔ͍Ͱ͍Δε λʔτΞοϓΛ໨ܸͨ͠ɻ

  4. confidencial 4 Զͷ໊લ͸ɺߴߍੜ୳ఁ޻౻৽Ұ AIΤϯδχΞͱͯ͠ಇ͖࢝ΊͨԶ͸ɺLLMΛ࢖ͬͯɺͨ͘͞Μ͓ۚΛՔ͍Ͱ͍Δε λʔτΞοϓΛ໨ܸͨ͠ɻ ʮͪΐͪΐͬͱ։ൃͨ͠ΒϘϩṶ͚Ͱ͖Δ΍Μʂʯͱؾ͍ͮͨԶ͸ɺىۀͯ͠ɺͻͨ ͢ΒPoCϓϩδΣΫτΛΫϥΠΞϯτʹఏڙ͢Δ͜ͱʹͨ͠ɻ

  5. confidencial 5 Զͷ໊લ͸ɺߴߍੜ୳ఁ޻౻৽Ұ AIΤϯδχΞͱͯ͠ಇ͖࢝ΊͨԶ͸ɺLLMΛ࢖ͬͯɺͨ͘͞Μ͓ۚΛՔ͍Ͱ͍Δε λʔτΞοϓΛ໨ܸͨ͠ɻ ʮͪΐͪΐͬͱ։ൃͨ͠ΒϘϩṶ͚Ͱ͖Δ΍Μʂʯͱؾ͍ͮͨԶ͸ɺىۀͯ͠ɺͻͨ ͢ΒPoCϓϩδΣΫτΛΫϥΠΞϯτʹఏڙ͢Δ͜ͱʹͨ͠ɻ Զ͸ɺഎޙ͔Β͍ۙͮͯ͘Δrate limit੍ݶͱɺLatencyͷѱԽʹؾ͕͍͍ͭͯͳ͔ͬ ͨɻ

  6. confidencial 6 Զͷ໊લ͸ɺߴߍੜ୳ఁ޻౻৽Ұ AIΤϯδχΞͱͯ͠ಇ͖࢝ΊͨԶ͸ɺLLMΛ࢖ͬͯɺͨ͘͞Μ͓ۚΛՔ͍Ͱ͍Δε λʔτΞοϓΛ໨ܸͨ͠ɻ ʮͪΐͪΐͬͱ։ൃͨ͠ΒϘϩṶ͚Ͱ͖Δ΍Μʂʯͱؾ͍ͮͨԶ͸ɺىۀͯ͠ɺͻͨ ͢ΒPoCϓϩδΣΫτΛΫϥΠΞϯτʹఏڙ͢Δ͜ͱʹͨ͠ɻ Զ͸ɺഎޙ͔Β͍ۙͮͯ͘Δrate limit੍ݶͱɺLatencyͷѱԽʹؾ͕͍͍ͭͯͳ͔ͬ ͨɻ

    ؾ͕͍ͭͨΒɺԶͷϓϩμΫτ͸ղ໿͕૬࣍͗ɺձࣾ͸౗࢈ͯ͠͠·͍ͬͯͨ...

  7. confidencial 7 ׬

  8. confidencial 8 ࠓ೔͸ɺ޻౻৽Ұ܅͕ɺ͜Μͳ݁຤Λܴ͑ͳ͍ͨΊʹͰ͖Δ͜ͱΛ͓࿩͠͠·͢ɻ

  9. confidencial ࣗݾ঺հ 2024/08 ೖࣾ SWEɾػցֶशΤϯδχΞͳͲΛܦݧ LLM͕ίΞʹͳΓͦ͏ͳαʔϏεͩͱࢥͬͯIVRyʹೖࣾ Moriya Hiroyuki 9 AI

    engineer

  10. confidencial IVRyͰͷLLMΛར༻ͨ͠AIର࿩ 10 WebsocketΛར༻͠ΤϯυϢʔβʔͱLLM͕ϦΞϧλΠϜʹ΍ΓऔΓ͍ͯ͠Δ

  11. confidencial LLM Fallback 11 ෳ਺ͷLLMΛར༻͢Δ͜ͱΛલఏʹFallbackػߏΛߏங APIͷStatus, Ratelimit΍σʔλ੍໿(஍ཧ੍໿)Λ΋ͱʹৼΓ෼͚

  12. confidencial LLM Fallback 12 ෳ਺ͷLLMΛར༻͢Δ͜ͱΛલఏʹFallbackػߏΛߏங APIͷStatus, Ratelimit΍σʔλ੍໿(஍ཧ੍໿)Λ΋ͱʹৼΓ෼͚ ؂ࢹ͢Ε͹ ྑ͍ͷ͡Ό

  13. confidencial ํ๏ 1ɿDataDog LLM observability 13 DataDog͕Ӷҙ։ൃதͷLLM؂ࢹʹಛԽͨ͠ػೳɻ Latency, token਺, promptͳͲΛऔಘͰ͖Δɻ

  14. confidencial 14 ʮ͜ΕͰɺOpenAIͷlatency͕؂ࢹͰ͖ΔΑ͏ʹͳͬͨͥʂʯ

  15. confidencial 15 ͋ΕΕʙɺ͓͔͍͠Αʙ

  16. confidencial 16 ͋ΕΕʙɺ͓͔͍͠Αʙ ๻ΒͷϓϩμΫτ͸ɺfallbackػߏΛ ࣮૷͍ͯ͠Δͷʹɺ OpenAIͷlatency͔͠؂ࢹͰ͖ͯͳ͍Αʙ

  17. confidencial ํ๏ 2ɿOpenLIT (OpenTelemetry) 17 OpenTelemetryن֨ʹଇͬͨɺLLM؂ࢹʹಛԽͨ͠πʔϧɻ ༷ʑͳLLMΛ؂ࢹ͢Δ͜ͱ͕Ͱ͖Δɻ

  18. confidencial 18 ʮ͜ΕͰɺ৭ʑͳmodelͷlatency͕؂ࢹͰ͖ΔΑ͏ʹͳͬͨͥʂʯ

  19. confidencial 19 ͋ΕΕʙɺ͓͔͍͠Αʙ

  20. confidencial 20 ͋ΕΕʙɺ͓͔͍͠Αʙ ๻Β͸ɺ৭ʑͳϞσϧΛ࢖͏Μ͔ͩΒɺ provider͝ͱʹɺlatencyΛܭଌ͢Δඞཁ͕͋Δͷʹ LiteLLMશମͰͷlatency͔͠औΕͯͳ͍Αʙ

  21. confidencial ํ๏ 3ɿDataDog Inferred services 21 DataDogʹ౥ࡌ͞ΕͨɺApp֎΁ͷϦΫΤετΛ؂ࢹͯ͘͠ΕΔػߏ

  22. confidencial 22 ʮ͜ΕͰɺLiteLLMͰ࢖͍ͬͯΔ͢΂ͯͷmodelΛ؂ࢹͰ͖ΔΑ͏ʹͳͬͨͥʂʯ

  23. confidencial 23 ͋ΕΕʙɺ͓͔͍͠Αʙ

  24. confidencial 24 ͋ΕΕʙɺ͓͔͍͠Αʙ ๻Β͸ɺGeminiɺOpenAIͰ̍ͭͷmodelΛ ࢖͏ͱ͸ݶΒͳ͍ͷʹɺ ݸผͷmodelͷlatencyΛऔಘ͢Δ͜ͱ͸ Ͱ͖ͯͳ͍Αʙ

  25. confidencial ·ͱΊ LLM؂ࢹ͸ɺ·ͩ·ͩൃల్্Ͱ஌ݟ͕͋Γ·ͤΜʂ AIɾLLMΛ࢖͍͜ͳͯ͠ϓϩμΫτʹೖΕ͍ͯ͘աఔͰɺ ࣗΒ͕੾Γ։͍͍ͯ͘ඞཁ͕͋Γ·͢ɻ ͥͻҰॹʹAI؂ࢹΛ΍͍͖ͬͯ·͠ΐ͏ʂ 25