DEV Community: Walnut Serv

Node demos for YouTube / Instagram / TikTok public data APIs (RapidAPI)

Walnut Serv — Fri, 24 Jul 2026 08:24:46 +0000

I published copy-paste Node demos for five public-data APIs on RapidAPI (free tier):

Repo: https://github.com/walnutserv-byte/beam-api-demos

Need	API	Hub
YouTube captions	CaptionBeam	https://rapidapi.com/WRT/api/captionbeam
IG profile + posts	InstaBeam	https://rapidapi.com/WRT/api/instabeam
TikTok user	TikBeam	https://rapidapi.com/WRT/api/tikbeam
Amazon ASIN	AmzBeam	https://rapidapi.com/WRT/api/amzbeam
Reddit subreddit	RedBeam	https://rapidapi.com/WRT/api/redbeam

Postman collection is in the repo (Beam-APIs.postman_collection.json).

const res = await fetch(
  "https://instabeam.p.rapidapi.com/profile?username=natgeo&count=6",
  {
    headers: {
      "X-RapidAPI-Key": process.env.RAPIDAPI_KEY,
      "X-RapidAPI-Host": "instabeam.p.rapidapi.com",
    },
  }
);
console.log(await res.json());

Built these because Graph App Review + brittle scrapers kept slowing side projects. Feedback welcome.

Bulk Instagram Profile & Posts API for Influencer Tools and Social Listening

Walnut Serv — Sat, 18 Jul 2026 05:32:21 +0000

If you're building anything that needs Instagram public data at scale — influencer discovery, social listening, competitor dashboards, content research bots — you already know the pain: Meta Graph App Review is a maze for public scraping use cases, unofficial endpoints flake overnight, and datacenter IPs get 429'd before your first 100 usernames.

This is a practical write-up on what builders actually need for bulk Instagram API calls, and a lightweight API we shipped for it: InstaBeam.

Who this is for

Teams enriching CRM / lead lists with Instagram follower counts and bios
Influencer marketplaces that need profile + recent posts without babysitting scrapers
Social listening / brand monitoring that polls public accounts on a schedule
Growth tools, RAG enrichment, and research scripts that just need clean JSON
Indie hackers who don't want to farm residential proxies or Instagram sessions

The pain (why DIY dies in production)

Meta Graph is built for "your own assets" — not "any public username I type in"
App Review + permissions burn weeks for use cases that feel obviously public
web_profile_info and similar internals rotate / schema-break without warning
Cloud egress IPs get blocked or rate-limited fast; residential proxies are their own ops job
Playwright + login farms scale worse than they look in a weekend demo

You don't want a headless browser fleet for "give me @natgeo followers + last 12 posts."

What "good enough" looks like for bulk

For most pipelines you need a boring HTTP shape:

Profile by username or profile URL
Follower / following / posts counts
Recent posts & reels (likes, comments, caption, media URL)
Single media by shortcode / permalink
Comments when Instagram still allows anonymous access

That's it. No session cookies. No Graph tokens. Cache by username.

A clean request shape

InstaBeam is on RapidAPI. Typical calls:

GET https://instabeam.p.rapidapi.com/profile?username=natgeo&count=12
X-RapidAPI-Key: YOUR_KEY
X-RapidAPI-Host: instabeam.p.rapidapi.com

GET https://instabeam.p.rapidapi.com/posts?username=natgeo&count=12
X-RapidAPI-Key: YOUR_KEY
X-RapidAPI-Host: instabeam.p.rapidapi.com

GET https://instabeam.p.rapidapi.com/media?shortcode=DZpQwxqimz2
X-RapidAPI-Key: YOUR_KEY
X-RapidAPI-Host: instabeam.p.rapidapi.com

Also: GET /health, GET /comments?shortcode=...

Bulk usage tips

Cache aggressively by username / shortcode (profiles don't change every second)
Batch usernames with backoff; don't hammer one account in a tight loop
Store id + shortcode so later jobs skip re-resolve
Prefer /posts when you only need the feed; /profile when you need counts + bio context
Start on the free tier to prove the pipeline, then jump plans when volume is real

Try it

API listing: https://rapidapi.com/WRT/api/instabeam

Built for nerds who need bulk Instagram profile/posts JSON without becoming a proxy company.

Written by Walnut Serv — InstaBeam Team

Bulk YouTube Transcripts for RAG, Summarizers, and Search Pipelines

Walnut Serv — Fri, 17 Jul 2026 20:01:39 +0000

Bulk YouTube Transcripts for RAG, Summarizers, and Search Pipelines

If you're building anything that needs spoken text from YouTube at scale — RAG indexes, content summarizers, caption tools, media search — you already know the pain: scraping captions yourself breaks, datacenter IPs get blocked, and rolling your own proxy stack eats weeks.

This is a practical write-up on what actually works for bulk YouTube transcript fetching, and a lightweight API we built for it: CaptionBeam.

Who this is for

Teams indexing large YouTube catalogs for search / RAG
Apps that summarize videos or generate show notes
Caption / subtitle tooling that needs reliable text + timestamps
Indie builders who don't want to babysit residential proxies

What usually goes wrong

No official public captions API for arbitrary videos
Cloud IPs get rate-limited or blocked fast
DIY with youtube-transcript libraries works in demos, then dies in production
Whisper-everywhere is expensive when captions already exist

For most bulk workloads, you want: captions first, speech-to-text only as fallback.

A clean request shape

CaptionBeam is hosted on RapidAPI. Typical call:

GET https://captionbeam.p.rapidapi.com/transcript?video_id=jNQXAC9IVRw&format=text
X-RapidAPI-Key: YOUR_KEY
X-RapidAPI-Host: captionbeam.p.rapidapi.com

Options:

video_id or full YouTube url
languages (comma-separated preference)
format: json | text | raw (timed cues)

Health check: GET /health

Bulk usage tips

Cache aggressively by video_id + language
Prefer format=text when you only need plain content for embeddings
Use raw when you need timestamps for clipping / alignment
Respect plan quotas; start free, scale when the pipeline is proven

Try it

API listing: https://rapidapi.com/WRT/api/captionbeam

Free tier is intentionally small so you can validate integration without committing. Paid plans are priced for builders doing real volume.

Written by Walnut Serv — CaptionBeam Team

I built Castella — a simple Android podcast player for offline listening

Walnut Serv — Fri, 26 Jun 2026 10:24:27 +0000

[Boost]

Walnut Serv — Tue, 23 Jun 2026 19:55:49 +0000

Walnut Serv

Jun 23

I built Castella — a simple Android podcast player for offline listening

#showdev #android #podcast #indie

1 min read

I built Castella — a simple Android podcast player for offline listening

Walnut Serv — Tue, 23 Jun 2026 18:51:19 +0000

I wanted a podcast app that gets out of the way: search a show, subscribe, download for offline, press play.

So I built Castella — an Android podcast player focused on daily listening.

What it does

Search podcasts and subscribe via RSS / directories
Download episodes for offline (commute, gym, flights)
Queue, favorites, and playback history
Speed control, sleep timer, chapters on supported shows
Background playback + lock-screen controls

Who it's for

Android listeners who want something simple and reliable — not another bloated media app.

Try it

Google Play: https://play.google.com/store/apps/details?id=com.wrt.castella

I'd love feedback from podcast listeners and Android devs. What would make you switch from your current player?

walnutserv@gmail.com

How to Fetch YouTube Transcripts for AI Summarization and RAG

Walnut Serv — Tue, 23 Jun 2026 17:35:26 +0000

If you're building AI apps that summarize YouTube videos, power RAG over video content, or generate subtitles, you need reliable transcript text — not browser scraping that breaks every other week.

This guide shows a simple REST approach that returns JSON with timestamps, plain text, or raw timed cues.

Why transcripts matter for AI

Summarization — feed transcript text to GPT/Claude instead of sending video
RAG — chunk timed cues into a vector database for semantic search
Accessibility — build caption tools without manual SRT editing
Content indexing — search across a channel's spoken content

Quick start with curl

curl "https://get-youtube-transcript.p.rapidapi.com/transcript?video_id=jNQXAC9IVRw&format=json" \
  -H "X-RapidAPI-Key: YOUR_KEY" \
  -H "X-RapidAPI-Host: get-youtube-transcript.p.rapidapi.com"

Replace YOUR_KEY with a key from the YouTube Transcript API on RapidAPI. The Basic plan includes 100 free requests/month.

Python example

import requests

API_URL = "https://get-youtube-transcript.p.rapidapi.com/transcript"
headers = {
    "X-RapidAPI-Key": "YOUR_KEY",
    "X-RapidAPI-Host": "get-youtube-transcript.p.rapidapi.com",
}
params = {"video_id": "jNQXAC9IVRw", "format": "json"}

data = requests.get(API_URL, headers=headers, params=params, timeout=60).json()

for cue in data["transcript"]:
    print(f"[{cue['start']:.1f}s] {cue['text']}")

Response formats

format	Use case
`json`	LLM pipelines, metadata + timestamps
`text`	Simple summarization
`raw`	Subtitle/SRT workflows

Parameters

video_id — 11-char YouTube ID or pass url with full YouTube link
languages — comma-separated codes, e.g. en,pt

Production tip

For batch jobs (indexing hundreds of videos), upgrade to the Ultra plan on RapidAPI — 100k requests/month at $9. Test in the playground first with any public video ID.

Disclosure: I built this API. Feedback welcome — especially on languages, latency, and batch endpoints.

Try it: https://rapidapi.com/wrt/api/get-youtube-transcript