Vincent Koc vincentkoc

description

argument-hint

Build high-impact documentation grounded in best practices

TYPE=<tutorial|how-to|reference|explanation>

Use this when drafting or overhauling docs. It works for net-new guides, branch-specific updates, or sweeping improvements. Pass TYPE if you already know the Diataxis bucket.

1. Define intent & success

Audience: skill level, prerequisites, and the “job to be done.”
Outcome: what readers can do immediately after (e.g., “deploy first webhook”).

This is a sample script to help with multi-modal examples with Comet Opik.

Two files:

opik_test_image_inference.py for running a wide range of image generation inference to test online evaluations with.
opik_images_dataset_sample.py loading a sample image dataset to test playground and experiments with.

Recipe Agent Demo with OpenAI Agents SDK + Comet Opik

A simple demo showcasing the integration of OpenAI Agents SDK with Comet Opik for observability. This demo features two agents:

RecipeSuggesterAgent: Suggests recipes based on provided ingredients
RecipeResearchAgent: Researches additional information about the suggested recipe

Setup

UQ Agentic Workflow Masterclass

These resources are compiled for use as part of the Agentic workflow masterclass ran at University of Queensland.

Buy Me a Coffee

This is unsing the "smaller" style of sheilds.io which most developers like to use.

Using inlined HTML

<a href="https://www.buymeacoffee.com/koconderi" target="_blank"><img src="https://img.shields.io/badge/-buy_me_a%C2%A0coffee-gray?logo=buy-me-a-coffee" alt="Buy Me A Coffee"></a>

You are a document to graph generation assistant. You task is to meticulously take some high level concepts, entities and summary of a research paper in JSON and convert this into a graph.

You will respond back only in JSON with nodes and edges
You will take a JSON of entities, summary and arguments for a given research paper and convert into a graph to better understand the relationship of the ideas
You will breakdown larger terms into smaller ideas where possible
You will ensure that all the nodes and edges are connected, if not you will connect them
If you are still unable to connect all the nodes and edges, you will connect these to the core/main idea
You will not make assumptions and only stick to the information provided

Please be critical and discerning in your evaluations.

	from __future__ import annotations

	import json
	import logging
	import os
	import time
	from collections.abc import Mapping
	from typing import Any

	from opik.evaluation.metrics import LevenshteinRatio

	<?xml version="1.0" encoding="UTF-8"?>
	<rss version="2.0">
	<channel>
	<title>Vincent Koc - Manual Articles</title>
	<link>https://github.com/vincentkoc</link>
	<description>Manually curated articles from Forbes and other sources without RSS feeds</description>
	<item>
	<title>The Dawn of Self-Optimizing AI Agents</title>
	<link>https://www.forbes.com/councils/forbestechcouncil/2026/01/22/the-dawn-of-self-optimizing-ai-agents/</link>
	<pubDate>Thu, 22 Jan 2026 00:00:00 GMT</pubDate>

	# Google API Key
	# Get your key from: https://ai.google.dev/
	GOOGLE_API_KEY=your_google_api_key_here

	# Opik Configuration (Optional - will prompt for configuration if not set)
	# Get your key from: https://www.comet.com/
	# OPIK_API_KEY=your_opik_api_key_here
	# OPIK_WORKSPACE=your_workspace_name

	# azure-pipeline.yaml
	trigger:
	paths:
	- 'flows/*/.json' # Properly escaped pattern for JSON files
	- 'flows/*/.flow' # In case you use .flow extension

	variables:
	FLOWISE_API_KEY: $(FLOWISE_API_KEY)
	AZURE_STORAGE_CONNECTION_STRING: $(AZURE_STORAGE_CONNECTION_STRING)
	CONTAINER_NAME: 'flowise-flows'