> Markdown version of [/videos/1906-rag-s-not-dead-you-re-just-using-it-wrong-phil-nash](https://www.wearedevelopers.com/videos/1906-rag-s-not-dead-you-re-just-using-it-wrong-phil-nash). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md).

---

# RAG's Not Dead, You're Just Using It Wrong! - Phil Nash

Phil Nash proves RAG isn't dead—you are just relying on rigid wrappers. Discover how custom, agentic workflows finally conquer unstructured enterprise data like legacy PDFs.

- **Speakers:** Phil Nash
- **Event:** Coffee With Developers
- **Published:** June 17, 2026
- **Duration:** 28:20
- **URL:** https://www.wearedevelopers.com/videos/1906-rag-s-not-dead-you-re-just-using-it-wrong-phil-nash

## Summary

Despite industry claims that Retrieval-Augmented Generation is dead or easily replaced by standard API skills, the practice remains foundational for handling private, unstructured enterprise data. The real challenge lies in the assumption that RAG is a solved, one-size-fits-all capability. Because enterprise knowledge is heavily locked inside legacy print-focused formats, accurately extracting and structuring this information requires more than off-the-shelf wrappers. Parsing complex, layout-driven documents like PDFs continues to be a severe bottleneck for AI ingestion, necessitating specialized parsing pipelines to prevent garbage-in, garbage-out retrieval.

To solve this, IBM developed OpenRAG, a customizable, agentic RAG framework composed of robust open-source components. It utilizes Docling for intelligent document parsing, OpenSearch for license-free hybrid vector and keyword querying, and Langflow for visual workflow orchestration. Rather than following a rigid user-to-database query flow, agentic RAG empowers the LLM to autonomously decide when to search the database, ask follow-up questions, or trigger external developer tools. This modularity allows engineering teams to build highly customized conversational agents tailored to their specific verticals and proprietary datasets.

In practice, these advanced retrieval systems integrate seamlessly with unified communication APIs, such as Twilio ConversationRelay and Agent Connect, to power low-latency voice AI agents over phone and messaging channels. Beyond retrieval, large enterprises must also optimize their internal AI development tooling. Building internal coding assistants requires smart model routing—dynamically selecting the most cost-effective and capable model for a specific task rather than defaulting to the most expensive option. While AI drastically accelerates boilerplate generation, seasoned developers remain critical for debugging complex architecture and validating massive, AI-generated pull requests.

**Keywords:** RAG implementation challenges, agentic RAG frameworks, openrag orchestration, docling PDF parser, opensearch indexing, langflow visual builder, enterprise AI routing, proprietary data retrieval, hybrid vector search, twilio conversationrelay, low-latency voice AI, large legacy codebases, AI model orchestration, automated pull validation, open source AI infrastructure

## Chapters

1. **The state of retrieving augmented generation** (00:00) — Why retrieval-augmented generation remains a crucial and unsolved challenge for operating on custom domain knowledge.
1. **Introducing OpenRAG for custom data pipelines** (01:35) — How OpenRAG provides a baseline architecture for building highly customized generative agents tailored to unstructured data.
1. **The challenge of extracting data from print formats** (02:14) — Why legacy print formats present unique positional obstacles for modern data ingestion pipelines.
1. **Open source technology stack for workflow orchestration** (03:38) — How integrating specialized extraction tools alongside native search and drag-and-drop orchestrators establishes a robust data ingestion layer.
1. **Agentic search and mitigating data exposure risks** (05:19) — How autonomous agents dynamically query contextual databases to leverage proprietary enterprise data without exposing it to public training datasets.
1. **Exploring internal enterprise coding agents for legacy syntax** (09:33) — Deploying internal coding assistants specifically optimized to update large legacy code bases while managing context window usage costs.
1. **Strategies for curating reliable technology news** (12:25) — Practical advice for developers managing information overload by selecting verified tech newsletters instead of high-noise social comment sequences.
1. **Integrating knowledge bases with telephonic voice agents** (16:51) — Connecting vector databases to voice streams by leveraging real-time socket connections to handle transcription and response orchestration.
1. **Balancing open source access with enterprise monetization** (20:24) — How open source projects successfully balance full community access with intentional feature limitations for corporate and compliance environments.
1. **Evaluating generated code syntax and maintaining quality control** (22:24) — Why experienced developer oversight remains critical for debugging, optimizing, and approving automated code abstractions to prevent systemic architectural drift.

## Related Moments

- [Simplifying generative AI deployments using the RagStack opinionated framework](https://www.wearedevelopers.com/videos/966-accelerating-genai-development-harnessing-astra-db-vector-store-and-langflow-for-llm-powered-apps) (from "Accelerating GenAI Development: Harnessing Astra DB Vector Store and Langflow for LLM-Powered Apps")
- [Customizing enterprise intelligence via automated RAG and secure APIs](https://www.wearedevelopers.com/videos/1665-secure-and-private-ai-deepmask) (from "Secure and Private AI - DeepMask")
- [Understanding basic retrieval-augmented generation architectures in chatbots](https://www.wearedevelopers.com/videos/1130-chatbots-are-going-to-destroy-infrastructures-and-your-cloud-bills) (from "Chatbots are going to destroy infrastructures and your cloud bills")
- [Deploying the augmented chatbot on OpenShift AI platforms](https://www.wearedevelopers.com/videos/950-supercharge-your-cloud-native-applications-with-generative-ai) (from "Supercharge your cloud-native applications with Generative AI")
- [Adapting operational workflows and infrastructure for RAG systems](https://www.wearedevelopers.com/videos/1152-ai-model-management-life-circles-ml-ops-for-generative-ai-models-from-research-to-deployment) (from "AI Model Management Life Circles: ML Ops For Generative AI Models From Research to Deployment")
- [Scaling generative AI use cases across large enterprises](https://www.wearedevelopers.com/videos/916-beyond-the-hype-real-world-ai-strategies-panel) (from "Beyond the Hype: Real-World AI Strategies Panel")

## Related Articles

- [Graph and AI Trends 2026: Why Is AI Running but Not Yet Delivering?](https://www.wearedevelopers.com/magazine/680-graph-and-ai-trends-2026-why-is-ai-running-but-not-yet-delivering)
- [MLOps And AI Driven Development](https://www.wearedevelopers.com/magazine/82-mlops-and-ai-driven-development)
- [What is Agentic Programming and Why Should Developers Care?](https://www.wearedevelopers.com/magazine/625-what-is-agentic-programming-and-why-should-developers-care)
- [WWC24 Talk - Scott Hanselman - AI: Superhero or Supervillain?](https://www.wearedevelopers.com/magazine/469-wwc24-talk-scott-hanselman-ai-superhero-or-supervillain)

## Related Jobs

- [AI Software Engineer (Germany)](https://www.wearedevelopers.com/jobs/48317-ai-software-engineer-germany) at **Sunhat**
- [Software-Entwickler – RAG & Knowledgraph (m/w/d)](https://www.wearedevelopers.com/jobs/48330-software-entwickler-rag-knowledgraph-m-w-d) at **Riverty**
- [Principal Software Engineer, Enterprise AI Platform](https://www.wearedevelopers.com/jobs/ext/1467292-principal-software-engineer-enterprise-ai-platform) at **GitHub**
- [Staff Software Engineer, Copilot Experiences](https://www.wearedevelopers.com/jobs/ext/164361-staff-software-engineer-copilot-experiences) at **GitHub**
- [Senior AI Agent Software Engineer (Go, Python) (m/f/x)](https://www.wearedevelopers.com/jobs/48277-senior-ai-agent-software-engineer-go-python-m-f-x) at **Dynatrace**
- [AI & Machine Learning Engineer (all genders)](https://www.wearedevelopers.com/jobs/48217-ai-machine-learning-engineer-all-genders) at **msg**