Athina giúp các nhà phát triển theo dõi và đánh giá các ứng dụng LLMs của họ trong quá trình sản xuất.
Thiết lập việc theo dõi và bắt đầu chạy đánh giá ngay hôm nay.
Thông tin liên hệ khác, hãy truy cập trang liên hệ với chúng tôi(https://cal.com/shiv-athina/30min)
Athina AI Tên công ty: Athina AI .
Liên kết giá của Athina AI: https://athina.ai/#pricing
Liên kết Linkedin Athina AI: https://www.linkedin.com/company/athina-ai
Liên kết Github Athina AI: https://github.com/athina-ai/athina-evals?utm_source=navbar&utm_medium=website
Bắt đầu
$0/tháng
Lên đến 10k naný, Phân tích và Insights
Giám sát
Tuỳ chỉnh
Tất cả những gì ở Bắt đầu, Nhiều chỗ ngồi trong nhóm, Phân loại chủ đề, Quản lý những lời nhắc
Đánh giá
Tuỳ chỉnh
Tất cả những gì ở Giám sát, Đánh giá tự động, Hỗ trợ cá nhân hóa
Doanh nghiệp
Tuỳ chỉnh
Bộ đánh giá tùy chỉnh, Tinh chỉnh fine, Tuân thủ SOC-2, Triển khai tự lưu trữ
Để biết mức giá mới nhất, vui lòng truy cập liên kết này: https://athina.ai/#pricing
Lắng nghe mạng xã hội
How to Use LLMs as Evaluators | TDE Workshop: Shiv Sakhuja
In this workshop, Shiv reviews LLM evaluators and how to use them for real-world applications. Shiv is a Co-founder @ Athina.ai (YC W23). Led by: Shiv Sakhuja Connect with Shiv LinkedIn: https://www.linkedin.com/in/shivsakhuja/ Athina AI: https://athina.ai/ Keep up with The Data Entrepreneurs! 🎥 YouTube: https://www.youtube.com/@TheDataEntrepreneurs 👉 Discord: https://discord.gg/RSqZbF9ygh 📰 Medium: https://medium.com/the-data-entrepreneurs 📅 Events: https://lu.ma/tde 🗞️ Newsletter: https://the-data-entrepreneurs.ck.page/profile Intro - 0:00 Athina AI - 1:30 Why do we need evals? - 2:28 Different types of evals - 5:20 Evaluation with labeled data - 7:10 Why LLMs can be used as evaluators - 9:34 Evaluation without labeled data - 12:54 Evaluating your retrieval - 19:16 Evaluating summaries - 22:47 Other evaluation techniques - 25:00 How to use evals in production - 25:38 Athina AI (SDK + Platform) - 26:38 Q&A - 28:26 Evals when starting from speech - 30:00 Cost and latency considerations - 34:09 Using "traditional" ML-based techniques - 37:21 Inference vs evaluation abilities of LLMs - 38:38 How to use these in CI/CD pipelines - 41:23
Why LLMs Can Do Evals (even if they failed at inference)
A clip from a recent workshop with Shiv Sakhuja, Co-founder at Athina AI (YC W23). Here, Shiv explains why an LLM that failed at inference can still be a good evaluator. 🎥 Full talk: https://youtu.be/JyocXRkiIcA Connect with Shiv LinkedIn: https://www.linkedin.com/in/shivsakhuja/ Athina AI: https://athina.ai/ Keep up with The Data Entrepreneurs! 🎥 YouTube: https://www.youtube.com/@TheDataEntrepreneurs 👉 Discord: https://discord.gg/RSqZbF9ygh 📰 Medium: https://medium.com/the-data-entrepreneurs 📅 Events: https://lu.ma/tde 🗞️ Newsletter: https://the-data-entrepreneurs.ck.page/profile
AI News for Jan 09, 2025
Welcome to 'The Automated Daily', your ultimate source for a streamlined and insightful daily news experience. Please support this podcast by checking out our sponsors: -Get $2,000 off the purchase of a Tesla product - https://ts.la/ron46932 Today's topics: -Stagehand web automation framework -VLC media player AI subtitles -Apple Intelligence scam concerns -Atelico AI Engine for gaming -Athina.ai workflow templates -Weco AI Functions platform -AI-generated code reliability issues -AI impact on global workforce -StoreLauncher Shopify store builder -Meta's AI character controversy -https://github.com/browserbase/stagehand -https://techcrunch.com/2025/01/09/vlc-tops-6-billion-downloads-previews-ai-generated-subtitles/ -https://www.crikey.com.au/2025/01/08/apple-new-artificial-intelligence-rewords-scam-messages-look-legitimate/ -https://atelico.studio/blog/cloud-ai-for-video-games-is-dead-on-arrival -https://app.athina.ai/flows/templates -https://www.aifunction.com/ -https://www.nuanced.dev/blog/the-reliability-gap -https://www.cnn.com/2025/01/08/business/ai-job-losses-by-2030-intl/index.html -https://storelauncher.app/ -https://www.washingtonpost.com/opinions/2025/01/08/meta-ai-bots-backlash-racist/ Subscribe to edition specific feeds: - Top news * Apple Podcast (https://apple.co/3PTvdUF) * Spotify (https://spoti.fi/3ZYXAW2) * RSS (https://bit.ly/the_automated_daily_news) - Tech news * Apple Podcast (https://apple.co/3RYWbg4) * Spotify (https://spoti.fi/3S089pG) * RSS (https://bit.ly/the_automated_daily_tech) - Hacker news * Apple Podcast (https://apple.co/48QWyzj) * Spotify (https://spoti.fi/45zD1kf) * RSS (https://bit.ly/the_automated_daily_hacker_news) - AI news * Apple Podcast (https://apple.co/3M6Tg1o) * Spotify (https://spoti.fi/3tzOfrz) * RSS (https://bit.ly/the_automated_daily_hackernews_ai) Visit our website at https://theautomateddaily.com/ Send feedback to feedback@theautomateddaily.com Youtube (https://www.youtube.com/@TheAutomatedDaily) LinkedIn (https://www.linkedin.com/in/the-automated-daily/) X (Twitter) (https://x.com/automated_daily)