---
title: "Dwelvin Morgan (@nivlewd1) | Hashnode"
description: "Solo Dev and creator of Social Craft Ai and Prompt Optimizer. Read the latest articles by Dwelvin Morgan on Hashnode."
url: "https://hashnode.com/@nivlewd1"
---

@nivlewd1

# Dwelvin Morgan

@nivlewd1 · Joined March 2026

Solo Dev and creator of Social Craft Ai and Prompt Optimizer

[Website](https://socialcraftai.app/)

### [Why I Chose Remotion + FFmpeg for Server-Side Video Rendering](https://socialprompt.hashnode.dev/why-i-chose-remotion-ffmpeg-for-server-side-video-rendering?utm_source=hashnode&utm_medium=feed)

Building a video creation platform, the "Video Studio," presented a significant technical challenge: how to enable users to generate high-quality videos directly from the platform. This required a rob

8 min read · [socialprompt.hashnode.dev](https://socialprompt.hashnode.dev/why-i-chose-remotion-ffmpeg-for-server-side-video-rendering?utm_source=hashnode&utm_medium=feed)

[May 20](https://hashnode.com/posts/why-i-chose-remotion-ffmpeg-for-server-side-video-rendering/6a0d695c7e4b2e77c06f21c0)

### [We need to stop treating Prompt Engineering like "dark magic" and start treating it like software testing.](https://socialprompt.hashnode.dev/we-need-to-stop-treating-prompt-engineering-like-dark-magic-and-start-treating-it-like-software-testing?utm_source=hashnode&utm_medium=feed)

Here's the scenario. You spend two hours brainstorming and manually crafting what you think is the perfect system prompt. You explicitly say: "Output strictly in JSON. Do not include markdown formatti

5 min read · [socialprompt.hashnode.dev](https://socialprompt.hashnode.dev/we-need-to-stop-treating-prompt-engineering-like-dark-magic-and-start-treating-it-like-software-testing?utm_source=hashnode&utm_medium=feed)

[Mar 23](https://hashnode.com/posts/we-need-to-stop-treating-prompt-engineering-like-dark-magic-and-start-treating-it-like-software-testing/69c1b11c30a9b81e3aab2318)

### [Reducing LLM API Costs by 43%: A Technical Deep-Dive into Intelligent Prompt Routing](https://socialprompt.hashnode.dev/reducing-llm-api-costs-by-43-a-technical-deep-dive-into-intelligent-prompt-routing?utm_source=hashnode&utm_medium=feed)

If you're building AI-powered applications, you've probably noticed your API bills climbing faster than your user growth. With frontier models like Claude Opus 4.5 ($5/$25 per 1M tokens) and GPT-5.2 Pro ($21/$168 per 1M tokens), even moderate usage c...

6 min read · [socialprompt.hashnode.dev](https://socialprompt.hashnode.dev/reducing-llm-api-costs-by-43-a-technical-deep-dive-into-intelligent-prompt-routing?utm_source=hashnode&utm_medium=feed)

[Mar 12](https://hashnode.com/posts/reducing-llm-api-costs-by-43-a-technical-deep-dive-into-intelligent-prompt-routing/69b27c30f49618851939cc3a)

### [Why your AI keeps ignoring your safety constraints (and how we fixed it by engineering "Intent")](https://socialprompt.hashnode.dev/why-your-ai-keeps-ignoring-your-safety-constraints-and-how-we-fixed-it-by-engineering-intent?utm_source=hashnode&utm_medium=feed)

If you’ve spent any time prompting LLMs, you’ve probably run into this frustrating scenario: You tell the AI to prioritize "safety, clarity, and conciseness." So, what happens when it has to choose between making a sentence clearer or making it safer...

4 min read · [socialprompt.hashnode.dev](https://socialprompt.hashnode.dev/why-your-ai-keeps-ignoring-your-safety-constraints-and-how-we-fixed-it-by-engineering-intent?utm_source=hashnode&utm_medium=feed)

[Mar 12](https://hashnode.com/posts/why-your-ai-keeps-ignoring-your-safety-constraints-and-how-we-fixed-it-by-engineering-intent/69b27c2fef971238c45ae3a4)

### [How We Achieved 91.94% Context Detection Accuracy Without Fine-Tuning](https://socialprompt.hashnode.dev/how-we-achieved-9194-context-detection-accuracy-without-fine-tuning?utm_source=hashnode&utm_medium=feed)

How We Achieved 91.94% Context Detection Accuracy Without Fine-Tuning The Problem When building Prompt Optimizer, we faced a critical challenge: how do you optimize prompts without knowing what the user is trying to do? A prompt for image generation ...

6 min read · [socialprompt.hashnode.dev](https://socialprompt.hashnode.dev/how-we-achieved-9194-context-detection-accuracy-without-fine-tuning?utm_source=hashnode&utm_medium=feed)

[Mar 12](https://hashnode.com/posts/how-we-achieved-9194-context-detection-accuracy-without-fine-tuning/69b27c2def971238c45ae3a2)
