Recsys Pipeline Architect
Design composable recommendation and feed pipelines: Source, Hydrator, Filter, Scorer, Selector, SideEffect stages, with runnable scaffolds in TypeScript, Go, or Python.
- What
- Design composable recommendation and feed pipelines: Source, Hydrator, Filter, Scorer, Selector, SideEffect stages, with runnable scaffolds in TypeScript, Go, or Python.
- Cost
- Free
- Needs
- Use "Recsys Pipeline Architect" with your Muse.
- Install
- Copy the installer prompt below into your Muse — your agent does the rest.
Curated by Skill Harbor: a recommendation pipeline architect skill for designing the plumbing around your ranking model, not the model itself. It uses a six-stage framework popularized by xAI's open-sourced For You algorithm: Source fetches candidates from one or more origins (parallel), Hydrator enriches each candidate with metadata (parallel), Filter drops what should never be shown like blocked, expired, or duplicate items (sequential), Scorer assigns scores (sequential, later scorers see earlier scores), Selector sorts and returns the top K (single op), and SideEffect caches served IDs, logs impressions, and emits events asynchronously without ever blocking the response. It explains why the order matters (know candidates before paying to enrich them, filter before scoring to save compute) and produces runnable scaffolds in TypeScript, Go, or Python for feeds, RAG rerankers, notification triage, search reranking, or ad ranking. By @affaan-m, listed here with credit to its creator. From the affaan-m/ECC repository (MIT); the six-stage pattern is attributed to xAI's For You algorithm (Apache 2.0). Honest caveats: pure architecture guidance, nothing to install; the scoring function is your responsibility, this is the plumbing around it; not for model architecture work or operating a deployed pipeline. Skill Harbor never reviews the code, review it yourself before use.
Version:
Install
Copy the install package below, then paste it into MuseThe install prompt below already includes the vetting steps: your agent follows the community checklist before installing anything with executable code. Want more?
Use "Recsys Pipeline Architect" with your Muse. Prerequisites: a ranking or recommendation problem to solve (the scoring function itself is yours to provide). Pure architecture guidance, nothing to install. 1. Open the skill: https://github.com/affaan-m/ECC/blob/main/skills/recsys-pipeline-architect/SKILL.md and copy the full SKILL.md text. 2. Paste it into a chat with Muse and add: "Design the pipeline for [your ranking problem] using the six stages: Source, Hydrator, Filter, Scorer, Selector, SideEffect." 3. Ask for the runnable scaffold: "Generate the scaffold in [TypeScript/Go/Python] with the stage boundaries explicit." Tip: define the Filter stage ruthlessly first; every candidate filtered early is compute you never spend. Safety: a skill is plain-text instructions; it runs nothing by itself. Review generated pipeline code for data handling and logging before deploying anywhere near real users.
Saved to your recent installs. Find it anytime on /connect.
Questions
How do I install a build?
Every product page includes a copy-paste install prompt. Paste it into your Muse and it sets the build up for you — no manual configuration.
Where does my money go?
Straight to the seller. Skill Harbor never processes payments: checkout happens on the seller’s own page, usually Stripe.
What does the ✓ next to a creator’s name mean?
It means we confirmed the identity of the person behind the listing. It says nothing about the code itself — always check a build before installing it.