DSPy Has a Free API You Should Know About

DSPy is a framework for programming — not prompting — language models. Instead of writing brittle prompts, you define modules that learn their own prompts through optimization.

Why DSPy Replaces Prompt Engineering

A team had 50 hand-crafted prompts that broke every time they switched models. DSPy automatically optimizes prompts for each model — no manual prompt engineering needed.

Key Features:

• Programmatic LM Calls — Define what you want, not how to prompt
• Automatic Optimization — Prompts are learned, not written
• Composable Modules — Chain, branch, and loop LM calls
• Evaluation-Driven — Optimize against your metrics
• Model-Agnostic — Works with any LLM provider

Quick Start

pip install dspy

import dspy

lm = dspy.LM("openai/gpt-4")
dspy.configure(lm=lm)

# Define a simple QA module
qa = dspy.ChainOfThought("question -> answer")
result = qa(question="What is the capital of France?")
print(result.answer)  # "Paris"

Custom Module

class Summarizer(dspy.Module):
    def __init__(self):
        self.summarize = dspy.ChainOfThought("document -> summary")
        self.extract_keywords = dspy.Predict("document -> keywords")

    def forward(self, document):
        summary = self.summarize(document=document)
        keywords = self.extract_keywords(document=document)
        return dspy.Prediction(summary=summary.summary, keywords=keywords.keywords)

Optimization

from dspy.teleprompt import BootstrapFewShot

optimizer = BootstrapFewShot(metric=my_metric)
optimized_module = optimizer.compile(my_module, trainset=train_data)

Why Choose DSPy

1. No prompt engineering — prompts are optimized automatically
2. Composable — build complex pipelines from simple modules
3. Model-portable — switch models without rewriting prompts

Check out DSPy docs to get started.

---

Building AI systems? Check out my Apify actors or email [e-mail] for data extraction.