• AI Engineering
  • Posts
  • Build a Machine Learning Model Using Just Natural Language

Build a Machine Learning Model Using Just Natural Language

.. with just a prompt

Author

Sumanth P
June 09, 2025

In today’s newsletter:

  • Build Machine Learning Models just using Natural Language

  • Search and get the LLM Ready Data for AI Agents

  • 100+ Fine-tuning notebooks for LLaMA, Qwen, DeepSeek, Gemma, and Phi, all in one place.

Reading time: 3 minutes.

Build Machine Learning models just using natural language!

Plexe AI lets you create machine learning models by describing them in plain language.

Simply explain what you want, and the AI-powered system builds a fully functional model through an automated agentic approach.

Here is why it's a game changer:

  • Natural Language Models: Define models in plain English.

  • AI Agent Team: Automates planning, coding, testing, and packaging.

  • One-Call Model Build: Create models with a single method.

  • Ray Support: Fast, distributed training and evaluation.

  • Smart Data Handling: Auto schema inference and data generation.

👉 It’s 100% open source. You can find the repo here → Plexe AI 

Search the web and get LLM ready data with just a few lines of Python code!

Firecrawl released the new search endpoint that lets you perform web searches, and scrape their full content in a single operation.

Key Features:

  • Choose specific output formats (markdown, HTML, links, screenshots)

  • Search the web with customizable parameters (language, country, etc.)

  • Optionally retrieve content from search results in various formats

  • Control the number of results and set timeouts

Check out the below code to get started.

Unsloth AI released this repository containing 100+ fine-tuning notebooks for LLaMA, Qwen, DeepSeek, Gemma, and Phi, all in one place.

Includes complete guides & examples for:

  • Use cases: Tool-calling, Classification, Synthetic data & more

  • End-to-end workflow: Data prep, training, running & saving models

  • BERT, TTS Vision models & more

  • Training methods like: GRPO, DPO, Continued Pretraining, SFT, Text Completion & more!

  • Llama, Qwen, DeepSeek, Gemma, Phi & more

It’s 100% Open Source

That’s a Wrap

That’s all for today. Thank you for reading today’s edition. See you in the next issue with more AI Engineering insights.

PS: We curate this AI Engineering content for free, and your support means everything. If you find value in what you read, consider sharing it with a friend or two.

Your feedback is valuable: If there’s a topic you’re stuck on or curious about, reply to this email. We’re building this for you, and your feedback helps shape what we send.

WORK WITH US

Looking to promote your company, product, or service to 120K+ AI developers? Get in touch today by replying to this email.