Send a personalised weekly tech digest from RSS feeds with OpenAI

Find Your AI Training Job
Chat with DigiNo Toucan and get matched to remote roles in one minute.

Newsletter operators and SaaS founders are drowning in tech noise but have no time to curate it themselves. This AI automation reads the web for them, filters by their exact interests, and lands a clean digest in their inbox every week without any manual effort.

What This Automation Does

Pulls articles daily from multiple tech RSS feeds and normalises titles, summaries, and publish dates into a consistent format
Converts each article into a vector embedding via OpenAI and stores it for fast semantic retrieval later in the week
Runs a weekly summarisation pass where an AI agent queries stored articles against the client's stated interests and selects the most relevant stories
Formats the final selection into a clean, readable email and delivers it automatically to the client's inbox via Gmail

Tools Used

n8n
OpenAI
Gmail

Work faster

Stop Typing and Start Speaking With Typeless

Typeless turns what you say into clean, formatted text in any app on Mac, Windows, Linux, iOS and Android, which it rates at 4x faster than typing. It strips filler words, keeps only your final phrasing when you change your mind mid-sentence, and works in 100+ languages. Emails, AI prompts, client messages and docs, done by voice.

Try Typeless →

Where to Get Hired for This Skill

On Contra, top freelancers across this stack have earned 96 combined verified reviews from real client projects.

Source: Contra freelancer search · refreshed 26 May 2026

Start Earning as a Freelancer on Contra

Contra is a commission-free professional network for independents. Browse live AI automation work and keep what you earn.

Join Contra Free →

How To Build It

Wire up RSS sources to a daily fetch

Configure a daily scheduled trigger to pull articles from a curated list of tech RSS feeds such as Wired, TechCrunch, and The Verge, so fresh content is ingested every 24 hours without manual intervention.

Normalise and chunk incoming article content

Process each fetched article to extract its title, summary, and publish date into a consistent structure, then split longer content into smaller chunks so the embedding model can handle them accurately.

Embed articles into an in-memory vector store

Pass each normalised article through the OpenAI embeddings model and store the resulting vectors in memory, enabling the weekly agent to retrieve semantically relevant stories rather than relying on keyword matching alone.

Run the weekly AI summarisation agent

Activate a separate weekly scheduled trigger that prompts an AI agent with the client's specific topic interests and a target story count, queries the vector store for the most relevant articles, and produces concise, newsworthy summaries.

Format and deliver the digest to the client's inbox

Convert the agent's output into clean, email-friendly markdown, then send it to the client's Gmail inbox on schedule so they receive a polished, branded newsletter without touching the workflow themselves.

Pitfalls

In-memory vector storage is wiped on every workflow restart or redeploy, meaning a server hiccup mid-week can silently erase the entire article backlog before the weekly summary runs. Migrating the client to a persistent vector database like Pinecone early in the engagement prevents data loss complaints.
OpenAI rate limits can throttle the embedding step when processing a large batch of RSS articles simultaneously, causing the daily ingestion job to fail partway through and leaving gaps in coverage that the AI agent will never know about.
Gmail OAuth tokens expire and require re-authorisation, which will silently halt newsletter delivery until someone manually refreshes the connection. Setting up a monitoring alert or a simple failure notification is essential for any client on a retainer.

FAQ

Can I build this without coding?

Yes. The entire workflow is assembled through a visual interface by connecting pre-built nodes for scheduling, RSS reading, OpenAI, and Gmail. You will need to understand how to configure API credentials and set trigger schedules, but no custom code is required to get a working newsletter running.

How long does it take?

A basic version delivering weekly summaries to a single inbox can be built and tested in two to four hours. Budget extra time if the client wants custom RSS sources, a persistent vector database, or branded email formatting.

What can I charge?

This workflow lends itself to a monthly retainer model where you cover initial setup, ongoing source curation, and periodic tuning of the summarisation prompts to keep the output relevant to the client's industry. Scope and retainer size depend on how many feeds you manage and how frequently the client wants adjustments.

Which tool is required vs optional?

OpenAI is required for both the embedding and summarisation steps. Gmail is the default delivery channel but can be swapped for any email provider or even a Telegram bot if the client prefers chat delivery. The in-memory vector store works fine for most clients but should be replaced with a persistent option if they need articles retained across restarts.

This is original DigiNo analysis. The underlying automation pattern is a community workflow template – view the original on n8n.