# Retrieval Projects at AI Tinkerers

> Canonical HTML: https://aitinkerers.org/technologies/retrieval
> Markdown URL: https://aitinkerers.org/technologies/retrieval.md
> Technology record last updated: 2026-03-09T22:32:25Z
> Generated: 2026-09-21T10:49:54Z

Retrieval-Augmented Generation (RAG) grounds Large Language Models (LLMs) in external, verified knowledge, injecting real-time data to boost accuracy and eliminate hallucinations.

Retrieval is now synonymous with RAG: a critical AI framework that connects a generative model to an authoritative, external knowledge base. The process is efficient: a user query triggers a semantic search against a vector database, retrieving the most relevant document chunks (embeddings). This retrieved context is then injected into the LLM's prompt, forcing the model to generate a response grounded in specific, up-to-date facts, not just its static training data. This mechanism dramatically improves factual accuracy, reduces the risk of AI hallucination by over 90% in some enterprise applications, and provides verifiable source citations.

- Official technology site: https://en.wikipedia.org/wiki/Retrieval-augmented_generation
- Public AI Tinkerers demos and talks: 4
- Result page: 1 of 1

## Recent Public Talks and Demos

### [Translating a book is a document problem, not a prompt problem](https://da-nang.aitinkerers.org/talks/rsvp_DOjIyS5U2Z8)

Translating a paragraph with an LLM is easy. Translating a real book is a document-understanding problem. Nga shares lessons from building Omniglot, an AI system for translating books and complex documents end to end. Omniglot processes structured PDFs, scans and images with challenging layouts, including writing systems with vertical and right-to-left reading order, before reconstructing documents into a representation suitable for translation. She will cover how the system combines document parsing, contextual translation memory, editor preferences and an embedded AI assistant that helps translators investigate terminology and compare translation choices. The talk explores a broader lesson from building applied AI products: the hardest problems often aren't the model call itself, but perception, context, memory, and designing AI that helps humans make better decisions. Nga Nguyen is the founder of Consen.AI and the builder of Omniglot. She works at the intersection of translation, machine learning and human-AI interaction, and is also a published Vietnamese translator. Omniglot was a Global Semifinalist at Microsoft Imagine Cup 2025.

- Event context: AI Tinkerers Da Nang: Inaugural Meetup — 2026-08-15 — Da Nang
- Public talk page: https://da-nang.aitinkerers.org/talks/rsvp_DOjIyS5U2Z8

### [From Cold Email to Trusted Conversation: How AI Agents Turn Outreach into Revenue](https://seattle.aitinkerers.org/talks/rsvp_Z5K4dyJQ7rE)

This talk covers the architecture of an AI-driven outbound system designed to create highly personalized B2B outreach at scale. I will walk through the design of two agent systems — IRIS, a hyper-personalized email generation engine, and Hermes, a social-channel warming system that leverages the psychological mere exposure effect to increase response rates. IRIS uses Firecrawl to extract structured information from target company websites, Perplexity for real-time research enrichment, and an LLM pipeline that synthesizes this context into personalized outreach messages rendered through React Email templates. The result is emails that read as if they were written by a human founder after researching the company. Hermes operates as a GUI-driven agent that manages warm-up sequences on social platforms, creating lightweight touchpoints before direct outreach. Together, these agents form a two-channel system designed to increase outbound effectiveness in trust-sensitive industries like healthcare. I will walk through the system architecture, data flow, and agent orchestration, including how contextual data is collected, synthesized, and translated into human-like messaging. I will also share early performance results showing how research-driven outreach can significantly outperform traditional cold email strategies.

- Event context: AI Tinkerers Seattle: GTM Track — March — 2026-03-26 — Seattle
- Public talk page: https://seattle.aitinkerers.org/talks/rsvp_Z5K4dyJQ7rE

### [Building an AI WhatsApp guide for the Valencia Fallas festival](https://valencia.aitinkerers.org/talks/rsvp_73Oz6MkCmvU)

I built an AI-powered WhatsApp assistant that acts as a digital guide for the Valencia Fallas festival. Visitors can ask about the main Fallas monuments and receive explanations about their meaning, satire, and artistic concept. For the most important Fallas, the assistant also delivers pre-recorded audio explanations in Spanish and Valencian, allowing visitors to experience them as if they were using an audio guide. The project demonstrates how conversational AI can turn a messaging app into an accessible cultural guide for large public events without requiring users to install a dedicated app.

- Event context: AI Tinkerers Valencia March Meetup — 2026-03-17 — Valencia
- Public talk page: https://valencia.aitinkerers.org/talks/rsvp_73Oz6MkCmvU

### [Oracle & NVIDIA AI‑Q: A Blueprint for High‑Performance Research Automation](https://singapore.aitinkerers.org/talks/rsvp_M4Vy0ySvk_Y)

Oracle & Nvidia will be presenting a technical deep-dive demo of an AI Research Assistant built using the NVIDIA AI‑Q blueprint approach, implemented with an enterprise-grade Oracle + NVIDIA stack—using Oracle Database 26ai as the system of record for vectors and retrieval, and NVIDIA’s accelerated AI software for embedding, retrieval optimization, and inference. At a high level, this presentation shows how to build and run a production-ready Retrieval-Augmented Generation (RAG) + agentic workflow where: Enterprise documents are ingested and embedded, using NVIDIA’s AI software stack (including NIM microservices and retrieval components). Embeddings are stored directly inside Oracle Database 26ai using its native VECTOR data type—so we don’t need a separate vector database. Semantic retrieval happens in Oracle Database 26ai using SQL (AI Vector Search), enabling vector similarity search combined with enterprise relational filters and governance. The retrieved context is then fed into a reasoning model to generate grounded answers and structured insights. What the Audience Will Learn / Take Away 1) How Oracle + NVIDIA changes enterprise AI architecture NVIDIA AI Enterprise is available natively through the OCI Console, reducing friction in provisioning AI software and accelerating adoption. Oracle and NVIDIA are co‑engineering deeper integrations, including NVIDIA NIM microservices support and NeMo Retriever integration with Oracle Database 26ai, enabling smoother RAG pipelines. 2) Why Oracle Database 26ai is a key differentiator for RAG Instead of deploying a separate vector DB, I’ll show how Oracle Database 26ai provides AI Vector Search directly inside the database, using the VECTOR data type—letting teams store embeddings next to business data and query semantically in SQL. 3) How to build a faster, simpler, more governable RAG workflow The demo will highlight architectural improvements that enterprise teams care about: - fewer moving parts, - less data duplication, - simpler security and governance patterns, - and a clean operational model that aligns with existing Oracle enterprise data platforms. What we’ll Show in the Demo (Step-by-Step) - Document ingestion (technical PDFs and enterprise materials) - Embedding generation using NVIDIA-optimized AI tooling (NIM microservices / retrieval stack). - Semantic retrieval using Oracle AI Vector Search in SQL (top‑K similarity search + filters). - Answer generation using reasoning over retrieved context

- Event context: AI Tinkerers - The Age of AI & Infrastructure (Singapore) — 2026-02-11 — Singapore
- Public talk page: https://singapore.aitinkerers.org/talks/rsvp_M4Vy0ySvk_Y

## Related Technologies

- [OpenAI](https://aitinkerers.org/technologies/openai) ([Markdown](https://aitinkerers.org/technologies/openai.md)) — 112 public demos
- [OpenAI API](https://aitinkerers.org/technologies/openai-api) ([Markdown](https://aitinkerers.org/technologies/openai-api.md)) — 520 public demos
- [Data](https://aitinkerers.org/technologies/data) ([Markdown](https://aitinkerers.org/technologies/data.md)) — 8 public demos
- [Document parsing](https://aitinkerers.org/technologies/document-parsing) ([Markdown](https://aitinkerers.org/technologies/document-parsing.md)) — 1 public demo
- [Embedding](https://aitinkerers.org/technologies/embedding) ([Markdown](https://aitinkerers.org/technologies/embedding.md)) — 2 public demos
- [Embeddings](https://aitinkerers.org/technologies/embeddings) ([Markdown](https://aitinkerers.org/technologies/embeddings.md)) — 22 public demos
- [Firecrawl](https://aitinkerers.org/technologies/firecrawl) ([Markdown](https://aitinkerers.org/technologies/firecrawl.md)) — 3 public demos
- [Image](https://aitinkerers.org/technologies/image) ([Markdown](https://aitinkerers.org/technologies/image.md)) — 4 public demos
- [Images](https://aitinkerers.org/technologies/images) ([Markdown](https://aitinkerers.org/technologies/images.md)) — 2 public demos
- [Inference](https://aitinkerers.org/technologies/inference) ([Markdown](https://aitinkerers.org/technologies/inference.md)) — 9 public demos
- [Infrastructure](https://aitinkerers.org/technologies/infrastructure) ([Markdown](https://aitinkerers.org/technologies/infrastructure.md)) — 1 public demo
- [LLM](https://aitinkerers.org/technologies/llm) ([Markdown](https://aitinkerers.org/technologies/llm.md)) — 123 public demos
- [LLM APIs](https://aitinkerers.org/technologies/llm-apis) ([Markdown](https://aitinkerers.org/technologies/llm-apis.md)) — 5 public demos
- [LLM translation](https://aitinkerers.org/technologies/llm-translation) ([Markdown](https://aitinkerers.org/technologies/llm-translation.md)) — 1 public demo
- [Multimodal Models](https://aitinkerers.org/technologies/multimodal-models) ([Markdown](https://aitinkerers.org/technologies/multimodal-models.md)) — 6 public demos
- [NeMo Retriever](https://aitinkerers.org/technologies/nemo-retriever) ([Markdown](https://aitinkerers.org/technologies/nemo-retriever.md)) — 1 public demo
- [NVIDIA AI Enterprise](https://aitinkerers.org/technologies/nvidia-ai-enterprise) ([Markdown](https://aitinkerers.org/technologies/nvidia-ai-enterprise.md)) — 1 public demo
- [OCR](https://aitinkerers.org/technologies/ocr) ([Markdown](https://aitinkerers.org/technologies/ocr.md)) — 10 public demos