Members-Only
Recent Talks & Demos are for members only
You must be an AI Tinkerers active member to view these talks and demos.
Tender Agent: RAG in Production
See a production system automating tender analysis and offer generation via scraping, RAG, vector databases, and prompt engineering, delivering clear business value.
Manually analyzing tender documents is a soul-crushing task for businesses. In this session, I’ll showcase our solution system we built to automate it. I’ll walk through our Tender analysis and Offer Generation modules, and show how they generate value instead of loads of text.
- Google GeminiGemini is Google's most capable, multimodal AI model: it seamlessly reasons across text, code, audio, image, and video.Gemini is Google's foundational, multimodal AI model, engineered to natively understand and combine text, code, image, audio, and video inputs. The technology is optimized across three sizes: Ultra (for highly complex tasks), Pro (for broad task scaling), and Nano (for efficient on-device performance). Gemini Ultra, for example, achieved a 90.0% score on the MMLU benchmark, surpassing human experts. It functions as a powerful AI assistant, integrated across Google services like Gmail and Maps, and features advanced tools like Deep Research and custom AI experts (Gems). Its Pro version offers a long context window, handling up to 1,500 pages or 30k lines of code simultaneously.
- gemini-embedding-001Gemini-embedding-001 is Google's premier, multilingual text embedding model, delivering state-of-the-art vector representations for semantic search and RAG systems.This is the flagship text embedding model for the Gemini API and Vertex AI, consistently holding a top spot on the Massive Text Embedding Benchmark (MTEB) Multilingual leaderboard. The model supports over 100 languages and accepts inputs up to 2,048 tokens. It generates dense vector representations (default 3072 dimensions) and utilizes Matryoshka Representation Learning (MRL), allowing developers to scale the output to smaller dimensions (e.g., 1536 or 768) for optimized storage and efficiency. Key applications include high-precision semantic search, text classification, clustering, and enhancing Retrieval Augmented Generation (RAG) systems.
- ScrapyScrapy is the high-level, open-source Python framework for fast, scalable web crawling and structured data extraction.Scrapy is a powerful, asynchronous application framework for web scraping and general-purpose web crawling (using 'Spiders'). Built on Python and the Twisted networking engine, it handles concurrent requests efficiently: it doesn't wait for a request to finish before sending the next one. This architecture ensures high performance and scalability for large-scale data projects. Developers use CSS selectors or XPath expressions for precise data extraction, then utilize built-in components like Item Pipelines for post-processing and Feed Exports to output scraped data directly to formats like JSON, CSV, or XML.
- KNIMEKNIME (Konstanz Information Miner) is the leading open-source platform for visual, low-code data science, integrating machine learning and ETL into drag-and-drop workflows.KNIME Analytics Platform is your open-source solution for end-to-end data science: it’s visual, intuitive, and highly effective. You build complete analytical pipelines—from data ingestion (ETL) to modeling (ML/AI)—by connecting modular 'nodes' in a 'workflow' . This drag-and-drop interface minimizes coding, making advanced analytics accessible to over 250,000 users globally . The platform is fully extensible: integrate your existing R and Python scripts directly, or leverage over 3,000 community-built components for specialized tasks . It’s a powerful, flexible tool, trusted by over 3,000 organizations across 40 industries .
Related projects
Multi-Agent System with Crew AI
Amman
Learn to build multi-agent systems with CrewAI. See agents collaborate on complex tasks, covering agents, roles, tasks, tools,…
Agentic AI at The Chefz: From LLMs to Business Impact
Amman
Learn how AI agents, built with n8n and GPT-4o, are practically deployed at The Chefz to proactively improve…
Agents using Advanced RAG Capabilities
Dubai
Demonstrates a proof‑of‑concept using modern retrieval methods—late interaction, OCR‑free pipelines, reasoning retrievers, and hypothetical embeddings—to improve agentic systems…
Tendermind- Tender Analyser
Hamburg
Demo of Tendermind, an LLM‑driven tender analyzer that extracts structured KPIs, uses Pydantic for JSON handling, and showcases…
AI science agent
Amsterdam
An experiment automating scientific research using GPT-5 Codex CLI resulted in a complete, non-trivial research paper, presented on…
Building Agentic AI for Intelligent Facility Operations
Toronto
Demonstration of an autonomous AI agent that ingests live building sensor data, identifies energy usage patterns, and automatically…