Content Synthesis Platform
Transform any source into structured content blocks. Deploy content recipes. All with one API call.
Recipes
Config
import DeepContent from "deepcontent";
const client = new DeepContent({ apiKey: "dc_your_key" });
const output = await client.run(
"deepcontent/linkedin-carousel:v1",
{ input: { source: "Today we shipped a new feature..." } }
);
console.log(output);
// => [{ type: "hook", content: "..." }, ...]Output preview
{
"status": "success",
"blocks": [
{
"type": "hook",
"content": "We just shipped..."
},
{
"type": "body",
"content": "Here's what changed..."
},
{
"type": "cta",
"content": "Try it now →"
}
],
"metadata": {
"tokens": 847,
"latency_ms": 1203
}
}Features
If you get a lot of traffic, DeepContent scales up automatically to handle the demand. If you don't get any traffic, we scale down to zero and don't charge you a thing.
DeepContent only bills you for how many syntheses your code is running. You don't pay for expensive GPUs when you're not using them.
Deploying content pipelines at scale is hard. If you tried, you need LLM orchestration, prompt management, caching, rate limiting, quality monitoring, output validation, and much more. DeepContent handles all of it.
Metrics let you keep an eye on how your recipes are performing and logs let you drill into particular syntheses to debug how your content is being generated.
How it works
You can get started with any content recipe using just one line of code. But as you do more complex things, you can fine-tune with your brand, or deploy your own custom recipes.
Our community has already published thousands of content recipes that are ready to use in production. You can run these with one line of code.
Explore recipesimport deepcontent
client = deepcontent.Client(api_token="dc_your_key")
output = client.run(
"deepcontent/linkedin-carousel:v1",
input={"source": "Today we shipped..."}
)
print(output)
# => [{"type": "hook", "content": "..."}, ...]You can customize recipes with your own brand voice and content patterns. Brand graphs encode your style. Recipe graphs encode your structure. Combined, they produce content that sounds exactly like you.
Create brand graph# Create a brand graph from your content
brand = client.brand_graphs.create(
name="Acme Corp Voice",
samples=[
"path/to/blog-post-1.md",
"path/to/blog-post-2.md",
"path/to/newsletter.md",
]
)
# Use it with any recipe
output = client.run(
"deepcontent/twitter-thread:v1",
input={
"source": "...",
"brand_graph": brand.id
}
)You are not limited to the recipes on DeepContent. Deploy your own custom recipes using JSON or our recipe DSL. We scale up and down to handle demand, and you only pay for the compute that you use.
Deploy a recipe{
"name": "custom-sales-email",
"version": "1.0",
"blocks": [
{ "type": "subject_line", "max_length": 60 },
{ "type": "hook", "style": "question" },
{ "type": "value_prop", "count": 3 },
{ "type": "social_proof", "source": "case_studies" },
{ "type": "cta", "urgency": "medium" }
],
"constraints": {
"tone": "professional",
"reading_level": "grade_8"
}
}Scale
Teams deploy content pipelines in a day and scale to millions of blocks, without hiring content operations staff.
Community
All the best content formats are on DeepContent. They are not just templates. They actually work and have production-ready structure. Why not contribute by publishing your own recipes?
Professional carousel with hook-value-cta structure
Engaging thread with numbered insights and callbacks
Weekly digest format with sections and highlights
Single-page sales narrative from product docs
Structured show notes with timestamps and quotes
Multiple ad variations optimized for different platforms