---
title: "Jun Bae (@jun07) | Hashnode"
description: "Developer. Read the latest articles by Jun Bae on Hashnode."
url: "https://hashnode.com/@jun07"
---

@jun07

# Jun Bae

@jun07 · Joined January 2026

Developer

### [Coroutine series 3) Coroutines for LLM inference](https://sjun.hashnode.dev/coroutine-series-3-coroutines-for-llm-inference?utm_source=hashnode&utm_medium=feed)

Learn how to utilize coroutines for LLM inference efficiently using asyncio and various AI frameworks like OpenAI, Google GenAI, and LangChain

13 min read · [sjun.hashnode.dev](https://sjun.hashnode.dev/coroutine-series-3-coroutines-for-llm-inference?utm_source=hashnode&utm_medium=feed)

[Feb 8](https://hashnode.com/posts/coroutine-series-3-coroutines-for-llm-inference/6988b15fe40ba20b660eaab5)

### [Coroutine series 2) Useful Asyncio Functions](https://sjun.hashnode.dev/coroutine-series-2-useful-asyncio-functions?utm_source=hashnode&utm_medium=feed)

Master Python asyncio for AI: Deep dive into TaskGroup vs. gather, structured concurrency, and using to\_thread for non-blocking I/O in production systems.

14 min read · [sjun.hashnode.dev](https://sjun.hashnode.dev/coroutine-series-2-useful-asyncio-functions?utm_source=hashnode&utm_medium=feed)

[Jan 31](https://hashnode.com/posts/coroutine-series-2-useful-asyncio-functions/697e2001ec23e8e4dbac4744)

### [Coroutine series 1) What are Coroutine, Asyncio, I/O bound?](https://sjun.hashnode.dev/1-what-are-coroutine-asyncio-io-bound?utm_source=hashnode&utm_medium=feed)

Explore Python coroutines and asyncio to efficiently handle I/O-bound tasks in AI applications

9 min read · [sjun.hashnode.dev](https://sjun.hashnode.dev/1-what-are-coroutine-asyncio-io-bound?utm_source=hashnode&utm_medium=feed)

[Jan 21](https://hashnode.com/posts/1-what-are-coroutine-asyncio-io-bound/6970fa4867fb984cf476e624)

### [KV Cache and Prompt Caching: How to Leverage them to Cut Time and Costs](https://sjun.hashnode.dev/kv-cache-and-prompt-caching-how-to-leverage-them-to-cut-time-and-costs?utm_source=hashnode&utm_medium=feed)

Learn how KV caching and prompt caching optimize LLM inference, reducing time and costs while balancing the challenge of increased memory consumption

10 min read · [sjun.hashnode.dev](https://sjun.hashnode.dev/kv-cache-and-prompt-caching-how-to-leverage-them-to-cut-time-and-costs?utm_source=hashnode&utm_medium=feed)

[Jan 16](https://hashnode.com/posts/kv-cache-and-prompt-caching-how-to-leverage-them-to-cut-time-and-costs/696a0290c49db289c77da82c)

### [Why LoRA? Understanding the representative PEFT.](https://sjun.hashnode.dev/why-lora?utm_source=hashnode&utm_medium=feed)

LoRA revolutionizes LLM fine-tuning, enabling efficient, cost-effective AI task adaptation without extensive resources

6 min read · [sjun.hashnode.dev](https://sjun.hashnode.dev/why-lora?utm_source=hashnode&utm_medium=feed)

[Jan 11](https://hashnode.com/posts/why-lora/6963cf0a97a4645cf6bed3d7)
