--- schema: 1 type: video title: 'DSPy: The End of Prompt Engineering - Kevin Madura, AlixPartners' description: >- Applications developed for the enterprise need to be rigorous, testable, and robust. The same is true for applications that use AI, but LLMs can make this challenging. In other words, you need to be able to program with LLMs, not just tweak prompts. In this talk we'll cover why DSPy really is all you need in building applications with LLMs. We'll dive into real-world examples where we have successfully automated manual work using an opinionated DSPy-first approach to structuring applications, covering everything from simple modules to using SoTA optimizers to measurably improve performance. https://x.com/kmad/ **Summary** Kevin Madura, a consultant at AlixPartners, argues that building robust enterprise AI applications requires shifting from brittle "prompt engineering" to "programming with LLMs" using **DSPy**. He contends that prompts should be treated as implementation details optimized by the system, while developers focus on defining typed interfaces (Signatures) and modular logic (Modules). The session moves from a conceptual overview of DSPy's primitives—Signatures, Modules, Adapters, and Optimizers—to a live code walkthrough. Madura demonstrates real-world use cases, including a complex pipeline that routes files by type (SEC filings vs. contracts) and a "boundary detector" that uses visual layout to segment legal documents. The talk concludes with a demonstration of how Optimizers (like MIPRO) can automatically tune these programs to outperform manual baselines, followed by a Q&A on production costs and feedback loops. **Timestamps** 00:00 Introduction & The Enterprise AI Challenge 07:12 The 6 Core Concepts of DSPy (Signatures, Modules, Adapters) 13:23 Deep Dive: Class-based vs. Shorthand Signatures 19:57 Adapters: Controlling the Prompt Format (JSON vs. BAML) 24:17 Optimizers: The "Killer Feature" for Transferability 31:08 Code Walkthrough: Setup & Model Mixing 36:24 Handling Documents: "Poor Man's RAG" with Attachments 42:10 Adapter Comparison: Improving Token Efficiency with BAML 47:20 Optimizers in Practice: Creating Datasets & Metrics 51:13 Complex Pipeline: Routing & Classifying Arbitrary Files 56:00 Advanced Use Case: PDF Boundary Detection via Visuals 01:01:22 Analyzing Optimization Results & The "DSPy Hub" Concept 01:09:02 Q&A: Handling Delayed Feedback & Online Learning 01:13:00 Conclusion --- Watch along with the corrected transcript, summary, timestamps and resources on this talk's official AIE page: https://ai.engineer/talks/-cKUW6n8hBU-dspy-the-end-of-prompt-engineering 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/channel/aidotengineer/explore/video/dspy-the-end-of-prompt-engineering-kevin-madura-ali-922e78aa2d76 markdown: >- https://liquidgold.cc/channel/aidotengineer/video/dspy-the-end-of-prompt-engineering-kevin-madura-ali-922e78aa2d76.md llms: null parent: 'https://liquidgold.cc/channel/aidotengineer.md' channel: aidotengineer updated: '2026-10-24T16:16:06.395Z' language: null children: [] videoId: '-cKUW6n8hBU' durationSeconds: 4393 publishedAt: '2026-01-08T20:48:21.000Z' searchChunks: - ordinal: 0 text: null startSeconds: 124 endSeconds: 226 watch: 'https://liquidgold.cc/watch/-cKUW6n8hBU?t=124' heldReason: rights - ordinal: 1 text: null startSeconds: 237 endSeconds: 329 watch: 'https://liquidgold.cc/watch/-cKUW6n8hBU?t=237' heldReason: rights - ordinal: 2 text: null startSeconds: 237 endSeconds: 329 watch: 'https://liquidgold.cc/watch/-cKUW6n8hBU?t=237' heldReason: rights - ordinal: 3 text: null startSeconds: 439 endSeconds: 451 watch: 'https://liquidgold.cc/watch/-cKUW6n8hBU?t=439' heldReason: rights - ordinal: 4 text: null startSeconds: 543 endSeconds: 670 watch: 'https://liquidgold.cc/watch/-cKUW6n8hBU?t=543' heldReason: rights - ordinal: 5 text: null startSeconds: 682 endSeconds: 812 watch: 'https://liquidgold.cc/watch/-cKUW6n8hBU?t=682' heldReason: rights - ordinal: 6 text: null startSeconds: 824 endSeconds: 905 watch: 'https://liquidgold.cc/watch/-cKUW6n8hBU?t=824' heldReason: rights - ordinal: 7 text: null startSeconds: 915 endSeconds: 998 watch: 'https://liquidgold.cc/watch/-cKUW6n8hBU?t=915' heldReason: rights - ordinal: 8 text: null startSeconds: 1146 endSeconds: 1204 watch: 'https://liquidgold.cc/watch/-cKUW6n8hBU?t=1146' heldReason: rights - ordinal: 9 text: null startSeconds: 1204 endSeconds: 1214 watch: 'https://liquidgold.cc/watch/-cKUW6n8hBU?t=1204' heldReason: rights - ordinal: 10 text: null startSeconds: 1214 endSeconds: 1462 watch: 'https://liquidgold.cc/watch/-cKUW6n8hBU?t=1214' heldReason: rights - ordinal: 11 text: null startSeconds: 1214 endSeconds: 1462 watch: 'https://liquidgold.cc/watch/-cKUW6n8hBU?t=1214' heldReason: rights - ordinal: 12 text: null startSeconds: 1214 endSeconds: 1462 watch: 'https://liquidgold.cc/watch/-cKUW6n8hBU?t=1214' heldReason: rights - ordinal: 13 text: null startSeconds: 1473 endSeconds: 1563 watch: 'https://liquidgold.cc/watch/-cKUW6n8hBU?t=1473' heldReason: rights - ordinal: 14 text: null startSeconds: 1473 endSeconds: 1563 watch: 'https://liquidgold.cc/watch/-cKUW6n8hBU?t=1473' heldReason: rights - ordinal: 15 text: null startSeconds: 1761 endSeconds: 1773 watch: 'https://liquidgold.cc/watch/-cKUW6n8hBU?t=1761' heldReason: rights - ordinal: 16 text: null startSeconds: 1887 endSeconds: 1969 watch: 'https://liquidgold.cc/watch/-cKUW6n8hBU?t=1887' heldReason: rights - ordinal: 17 text: null startSeconds: 1980 endSeconds: 2072 watch: 'https://liquidgold.cc/watch/-cKUW6n8hBU?t=1980' heldReason: rights - ordinal: 18 text: null startSeconds: 2858 endSeconds: 3081 watch: 'https://liquidgold.cc/watch/-cKUW6n8hBU?t=2858' heldReason: rights - ordinal: 19 text: null startSeconds: 3081 endSeconds: 3092 watch: 'https://liquidgold.cc/watch/-cKUW6n8hBU?t=3081' heldReason: rights - ordinal: 20 text: null startSeconds: 3092 endSeconds: 3361 watch: 'https://liquidgold.cc/watch/-cKUW6n8hBU?t=3092' heldReason: rights - ordinal: 21 text: null startSeconds: 3361 endSeconds: 3376 watch: 'https://liquidgold.cc/watch/-cKUW6n8hBU?t=3361' heldReason: rights - ordinal: 22 text: null startSeconds: 3701 endSeconds: 3781 watch: 'https://liquidgold.cc/watch/-cKUW6n8hBU?t=3701' heldReason: rights - ordinal: 23 text: null startSeconds: 3793 endSeconds: 3924 watch: 'https://liquidgold.cc/watch/-cKUW6n8hBU?t=3793' heldReason: rights - ordinal: 24 text: null startSeconds: 3935 endSeconds: 4024 watch: 'https://liquidgold.cc/watch/-cKUW6n8hBU?t=3935' heldReason: rights - ordinal: 25 text: null startSeconds: 4159 endSeconds: 4393 watch: 'https://liquidgold.cc/watch/-cKUW6n8hBU?t=4159' heldReason: rights --- # DSPy: The End of Prompt Engineering - Kevin Madura, AlixPartners Applications developed for the enterprise need to be rigorous, testable, and robust. The same is true for applications that use AI, but LLMs can make this challenging. In other words, you need to be able to program with LLMs, not just tweak prompts. In this talk we'll cover why DSPy really is all you need in building applications with LLMs. We'll dive into real-world examples where we have successfully automated manual work using an opinionated DSPy-first approach to structuring applications, covering everything from simple modules to using SoTA optimizers to measurably improve performance. https://x.com/kmad/ \*\*Summary\*\* Kevin Madura, a consultant at AlixPartners, argues that building robust enterprise AI applications requires shifting from brittle "prompt engineering" to "programming with LLMs" using \*\*DSPy\*\*. He contends that prompts should be treated as implementation details optimized by the system, while developers focus on defining typed interfaces (Signatures) and modular logic (Modules). The session moves from a conceptual overview of DSPy's primitives—Signatures, Modules, Adapters, and Optimizers—to a live code walkthrough. Madura demonstrates real-world use cases, including a complex pipeline that routes files by type (SEC filings vs. contracts) and a "boundary detector" that uses visual layout to segment legal documents. The talk concludes with a demonstration of how Optimizers (like MIPRO) can automatically tune these programs to outperform manual baselines, followed by a Q\&A on production costs and feedback loops. \*\*Timestamps\*\* 00:00 Introduction & The Enterprise AI Challenge 07:12 The 6 Core Concepts of DSPy (Signatures, Modules, Adapters) 13:23 Deep Dive: Class-based vs. Shorthand Signatures 19:57 Adapters: Controlling the Prompt Format (JSON vs. BAML) 24:17 Optimizers: The "Killer Feature" for Transferability 31:08 Code Walkthrough: Setup & Model Mixing 36:24 Handling Documents: "Poor Man's RAG" with Attachments 42:10 Adapter Comparison: Improving Token Efficiency with BAML 47:20 Optimizers in Practice: Creating Datasets & Metrics 51:13 Complex Pipeline: Routing & Classifying Arbitrary Files 56:00 Advanced Use Case: PDF Boundary Detection via Visuals 01:01:22 Analyzing Optimization Results & The "DSPy Hub" Concept 01:09:02 Q\&A: Handling Delayed Feedback & Online Learning 01:13:00 Conclusion --- Watch along with the corrected transcript, summary, timestamps and resources on this talk's official AIE page: https://ai.engineer/talks/-cKUW6n8hBU-dspy-the-end-of-prompt-engineering 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/channel/aidotengineer/explore/video/dspy-the-end-of-prompt-engineering-kevin-madura-ali-922e78aa2d76 · 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.