Gemini polish

.env.example

@@ -1,13 +1,22 @@
 #!/bin/bash
 # OpenAI-compatible LLM
-export OPENAI_API_BASE="https://api.openai.com/v1"
-export OPENAI_API_KEY="sk-proj-..."
+OPENAI_BASE_URL="https://generativelanguage.googleapis.com/v1beta/openai/"
+OPENAI_API_KEY="..."
+
+# Embedding model
+EMBEDDING_BASE_URL="https://..."
+EMBEDDING_API_KEY="..."
 
 # LangFuse
-export LANGFUSE_PUBLIC_KEY="pk-lf-..."
-export LANGFUSE_SECRET_KEY="sk-lf-..."
-export LANGFUSE_HOST="https://cloud.langfuse.com"
+LANGFUSE_SECRET_KEY="sk-lf-..."
+LANGFUSE_PUBLIC_KEY="pk-lf-..."
+LANGFUSE_HOST="https://us.cloud.langfuse.com"
 
-# ElasticSearch
-export ES_HOSTS="https://..."
-export ES_API_KEY="..."
+# Weaviate
+WEAVIATE_HTTP_HOST="...weaviate.cloud" # or 'localhost' for local Weaviate
+WEAVIATE_GRPC_HOST="grpc-...weaviate.cloud" # or 'localhost' for local Weaviate
+WEAVIATE_API_KEY="..."
+WEAVIATE_HTTP_PORT="443" # or 8080 for localhost
+WEAVIATE_GRPC_PORT="443" # or 50051 for localhost
+WEAVIATE_HTTP_SECURE="true" # set to false for localhost
+WEAVIATE_GRPC_SECURE="true" # set to false for localhost

.pre-commit-config.yaml

@@ -20,30 +20,21 @@ repos:
     - id: check-toml
 
   - repo: https://github.com/astral-sh/uv-pre-commit
-    rev: 0.7.2
+    rev: 0.7.19
     hooks:
       - id: uv-lock
 
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: 'v0.11.8'
+    rev: 'v0.12.2'
     hooks:
     - id: ruff
       args: [--fix, --exit-non-zero-on-fix]
       types_or: [python, jupyter]
     - id: ruff-format
       types_or: [python, jupyter]
 
-  - repo: https://github.com/pre-commit/mirrors-mypy
-    rev: v1.15.0
-    hooks:
-    - id: mypy
-      entry: python3 -m mypy --config-file pyproject.toml
-      language: system
-      types: [python]
-      exclude: "tests"
-
   - repo: https://github.com/crate-ci/typos
-    rev: v1  # v1.19.0
+    rev: v1.34.0
     hooks:
       - id: typos
         args: []
@@ -54,23 +45,6 @@ repos:
     - id: nbqa-ruff
       args: [--fix, --exit-non-zero-on-fix]
 
-  - repo: local
-    hooks:
-    - id: doctest
-      name: doctest
-      entry: python3 -m doctest -o NORMALIZE_WHITESPACE
-      files: "^aieng_template/"
-      language: system
-
-  # - repo: local
-  #   hooks:
-  #   - id: pytest
-  #     name: pytest
-  #     entry: python3 -m pytest -m "not integration_test"
-  #     language: system
-  #     pass_filenames: false
-      # always_run: true
-
 ci:
     autofix_commit_msg: |
         [pre-commit.ci] Add auto fixes from pre-commit.com hooks
@@ -80,5 +54,5 @@ ci:
     autoupdate_branch: ''
     autoupdate_commit_msg: '[pre-commit.ci] pre-commit autoupdate'
     autoupdate_schedule: weekly
-    skip: [pytest,doctest,mypy]
+    skip: [pytest,doctest]
     submodules: false

README.md

@@ -1,25 +1,148 @@
-# AI Engineering template (with uv)
+# Agent Bootcamp
 
 ----------------------------------------------------------------------------------------
 
-[![code checks](https://github.com/VectorInstitute/aieng-template-uv/actions/workflows/code_checks.yml/badge.svg)](https://github.com/VectorInstitute/aieng-template-uv/actions/workflows/code_checks.yml)
-[![integration tests](https://github.com/VectorInstitute/aieng-template-uv/actions/workflows/integration_tests.yml/badge.svg)](https://github.com/VectorInstitute/aieng-template-uv/actions/workflows/integration_tests.yml)
-[![docs](https://github.com/VectorInstitute/aieng-template-uv/actions/workflows/docs.yml/badge.svg)](https://github.com/VectorInstitute/aieng-template-uv/actions/workflows/docs.yml)
-[![codecov](https://codecov.io/github/VectorInstitute/aieng-template-uv/graph/badge.svg?token=83MYFZ3UPA)](https://codecov.io/github/VectorInstitute/aieng-template-uv)
-![GitHub License](https://img.shields.io/github/license/VectorInstitute/aieng-template-uv)
+This is a collection of reference implementations for Vector Institute's **Agent Bootcamp**, taking place between June and September 2025. The repository demonstrates modern agentic workflows for retrieval-augmented generation (RAG), evaluation, and orchestration using the latest Python tools and frameworks.
 
-A template repo for AI Engineering projects (using ``python``) and ``uv``. This
-template is like our original AI Engineering [template](https://github.com/VectorInstitute/aieng-template),
-however, unlike how that template uses poetry, this one uses uv for dependency
-management (as well as packaging and publishing).
+## Reference Implementations
 
-## 🧑🏿‍💻 Developing
+This repository includes several modules, each showcasing a different aspect of agent-based RAG systems:
 
-### Installing dependencies
+**1. Basics: Reason-and-Act RAG**
+A minimal Reason-and-Act (ReAct) agent for knowledge retrieval, implemented without any agent framework.
 
-The development environment can be set up using
-[uv](https://github.com/astral-sh/uv?tab=readme-ov-file#installation). Hence, make sure it is
-installed and then run:
+- **[1.0 Search Demo](src/1_basics/0_search_demo/README.md)**
+  A simple demo showing the capabilities (and limitations) of a knowledgebase search.
+
+
+- **[1.1 ReAct Agent for RAG](src/1_basics/1_react_rag/README.md)**
+  Basic ReAct agent for step-by-step retrieval and answer generation.
+
+**2. Frameworks: OpenAI Agents SDK**
+  Showcases the use of the OpenAI agents SDK to reduce boilerplate and improve readability.
+
+- **[2.1 ReAct Agent for RAG - OpenAI SDK](src/2_frameworks/1_react_rag/README.md)**
+  Implements the same Reason-and-Act agent using the high-level abstractions provided by the OpenAI Agents SDK. This approach reduces boilerplate and improves readability.
+  The use of langfuse for making the agent less of a black-box is also introduced in this module.
+
+- **[2.2 Multi-agent Setup for Deep Research](src/2_frameworks/2_multi_agent/README.md)**
+  Demo of a multi-agent architecture with planner, researcher, and writer agents collaborating on complex queries.
+
+**3. Evals: Automated Evaluation Pipelines**
+  Contains scripts and utilities for evaluating agent performance using LLM-as-a-judge and synthetic data generation. Includes tools for uploading datasets, running evaluations, and integrating with [Langfuse](https://langfuse.com/) for traceability.
+
+- **[3.1 LLM-as-a-Judge](src/3_evals/1_llm_judge/README.md)**
+  Automated evaluation pipelines using LLM-as-a-judge with Langfuse integration.
+
+- **[3.2 Evaluation on Synthetic Dataset](src/3_evals/2_synthetic_data/README.md)**
+  Showcases the generation of synthetic evaluation data for testing agents.
+
+
+# Reference Implementation Repository for AI Agent Bootcamp
+
+## Setting
+
+Build a knowledge retrieval agent.
+
+## Symposis
+
+**Basics**: Build a ReAct-styled ("Reason-and-Act") agent without using agent frameworks.
+
+**Frameworks**: Build agents with frameworks to reduce boilerplate code.
+
+**Evals**: Agent observability, test data synthesis, and automated evaluation.
+
+## Tooling
+
+- **openai-agents** as the AI Agent framework.
+- **langfuse** for agent observability and evaluation.
+- **uv** for dependency management
+- **gradio** for an interactive prototype.
+- **weaviate** as the search engine for the local knowledge base.
+
+
+## Getting Started
+
+Set your API keys in `.env`. Use `.env.example` as a template.
+
+```bash
+cp -v .env.example .env
+```
+
+Run integration tests to validate that your API keys are set up correctly.
+
+```bash
+PYTHONPATH="." uv run pytest -sv tests/tool_tests/test_integration.py
+```
+
+## Reference Implementations
+
+### 1. Basics
+
+Interactive knowledge base demo. Access the gradio interface in your browser (see forwarded ports.)
+
+```bash
+uv run --env-file .env -m src.1_basics.0_search_demo.gradio
+```
+
+Basic Reason-and-Act Agent- command line version. To exit, press `Control-\`.
+
+```bash
+uv run --env-file .env -m src.1_basics.1_react_rag.main
+```
+
+Interactive web version of the Gradio Reason-and-Act Agent.
+
+```bash
+uv run --env-file .env -m src.1_basics.1_react_rag.gradio
+```
+
+
+### 2. Frameworks
+
+Reason-and-Act Agent without the boilerplate- using the OpenAI Agent SDK.
+
+```bash
+uv run --env-file .env -m src.2_frameworks.1_react_rag.basic
+uv run --env-file .env -m src.2_frameworks.1_react_rag.gradio
+uv run --env-file .env -m src.2_frameworks.1_react_rag.langfuse_gradio
+```
+
+Multi-agent examples, also via the OpenAI Agent SDK.
+
+```bash
+uv run --env-file .env -m src.2_frameworks.2_multi_agent.gradio
+```
+
+### 3. Evals
+
+Synthetic data.
+
+```bash
+uv run -m src.3_evals.2_synthetic_data.synthesize_data \
+--source_dataset hf://vector-institute/hotpotqa@d997ecf:train \
+--langfuse_dataset_name search-dataset-synthetic-20250609 \
+--limit 18
+```
+
+Run LLM-as-a-judge Evaluation on synthetic data
+
+```bash
+uv run \
+--env-file .env \
+-m src.3_evals.1_llm_judge.run_eval \
+--langfuse_dataset_name search-dataset-synthetic-20250609 \
+--run_name enwiki_weaviate \
+--limit 18
+```
+
+## Requirements
+
+- Python 3.12+
+
+## Getting Started
+
+Clone the repository:
 
 ```bash
 uv sync

docs/index.md

@@ -17,5 +17,5 @@ Build a knowledge retrieval agent.
 - **openai-agents** as the AI Agent framework.
 - **langfuse** for agent observability and evaluation.
 - **uv** for dependency management
-- **streamlit** for an interactive prototype.
-- **elasticsearch** as the search engine for the local knowledge base.
+- **gradio** for an interactive prototype.
+- **weaviate** as the search engine for the local knowledge base.

pyproject.toml

@@ -5,58 +5,55 @@ description = "Add your description here"
 readme = "README.md"
 authors = [ {name = "Vector AI Engineering", email = "ai_engineering@vectorinstitute.ai"}]
 license = "Apache-2.0"
-repository = "https://github.com/VectorInstitute/aieng-template-uv"
+repository = "https://github.com/VectorInstitute/agent-bootcamp"
 requires-python = ">=3.12"
 dependencies = [
-    "elasticsearch[async]>=9.0.1",
-    "gradio>=5.31.0",
-    "openai>=1.82.0",
-    "openai-agents>=0.0.16",
-    "pydantic>=2.11.5",
+    "beautifulsoup4>=4.13.4",
+    "datasets>=3.6.0",
+    "gradio>=5.35.0",
+    "langfuse>=3.1.3",
+    "lxml>=6.0.0",
+    "nest-asyncio>=1.6.0",
+    "numpy<2.3.0",
+    "openai>=1.93.1",
+    "openai-agents>=0.1.0",
+    "pydantic>=2.11.7",
+    "pydantic-ai-slim[logfire]>=0.3.7",
     "pytest-asyncio>=0.25.2",
+    "weaviate-client>=4.15.4",
 ]
 
 [dependency-groups]
 dev = [
     "codecov>=2.1.13",
-    "mypy>=1.14.1",
+    "ipykernel>=6.29.5",
+    "ipython>=9.4.0",
+    "ipywidgets>=8.1.7",
+    "jupyter>=1.1.1",
+    "jupyterlab>=4.4.2",
     "nbqa>=1.9.1",
     "pip-audit>=2.7.3",
     "pre-commit>=4.1.0",
     "pytest>=8.3.4",
     "pytest-asyncio>=0.25.2",
     "pytest-cov>=6.0.0",
     "pytest-mock>=3.14.0",
-    "ruff>=0.9.2",
+    "ruff>=0.12.2",
 ]
 docs = [
     "jinja2>=3.1.6", # Pinning version to address vulnerability GHSA-cpwx-vrp4-4pq7
     "mkdocs>=1.6.0",
-    "mkdocs-material>=9.5.15",
+    "mkdocs-material>=9.6.15",
     "mkdocstrings>=0.24.1",
-    "mkdocstrings-python>=1.10.0",
+    "mkdocstrings-python>=1.16.12",
     "ipykernel>=6.29.5",
-    "ipython>=8.31.0",
+    "ipython>=9.4.0",
 ]
 
 # Default dependency groups to be installed
 [tool.uv]
 default-groups = ["dev", "docs"]
 
-[tool.mypy]
-install_types = true
-pretty = true
-namespace_packages = true
-explicit_package_bases = true
-warn_unused_configs = true
-allow_subclassing_any = false
-check_untyped_defs = true
-allow_untyped_decorators = false
-warn_redundant_casts = true
-warn_return_any = true
-strict_equality = true
-
-
 [tool.ruff]
 include = ["*.py", "pyproject.toml", "*.ipynb"]
 line-length = 88
@@ -95,6 +92,7 @@ ignore = [
     "PLR2004", # Replace magic number with named constant
     "PLR0913", # Too many arguments
     "COM812", # Missing trailing comma
+    "N999", # Number in module names.
 ]
 
 # Ignore import violations in all `__init__.py` files.

src/1_basics/0_search_demo/README.md

@@ -0,0 +1,10 @@
+# Knowledgebase Demo
+
+This folder contains logic for showcasing the capabilities (and limitations) of the knowledgebase.
+
+Format of the output is similar to what the Agent LLM will receive as tool output.
+
+```bash
+source .env && \
+uv run -m src.1_basics.0_search_demo.gradio
+```