YouTube Prompt Engineer & Script Quality Controller (AI Biblical History Network)

Please login or register as jobseeker to apply for this job.

TYPE OF WORK

Gig

WAGE / SALARY

$10-$15 per hour

HOURS PER WEEK

15

DATE UPDATED

Jun 13, 2026

JOB OVERVIEW

Channel Niche Notice: We run a high-volume media network deeply rooted in Biblical history, archaeology, and ancient Near Eastern narratives, with active plans to expand into broader historical eras (the Middle Ages, Roman history, etc.). A strong familiarity with, or a deep interest in, Biblical texts, theology, and ancient historical analysis is highly advantageous for this role.

Job Description:
PLEASE READ THIS FIRST: We are NOT looking for a traditional creative scriptwriter. We do not want someone to sit down at a blank page and type essays from scratch. That slows down our pipeline.

Instead, we are looking for a highly analytical Prompt Engineer & Script Quality Controller to oversee our automated historical media pipeline.

Our Content Factory utilizes advanced AI systems to handle high-volume script generation and deep historical research extraction. Your core mission is two-fold:

Front-End Engineering: You will actively develop, test, and write advanced system and API prompts to ensure our AI engines output structurally elite first drafts.

Back-End Quality Control: Once the scripts are generated by your prompts, you will act as the human firewall—checking, refining, and verifying the final text to ensure it strictly meets our high narrative and accuracy standards before going to production.

Please note: To prove you are reading carefully, you must include the word ANCIENT as the final word of your application message.

Key Responsibilities:

Prompt Optimization: Write and iterate complex system/API prompts (using frameworks like Claude and OpenAI) to force the AI to natively output specific audience retention mechanics (psychological open loops, micro-hooks every 45 seconds, cliffhangers, and perfect narrative pacing).

Output Quality Control: Methodically audit the AI-generated script outputs. If the AI text drops in quality, you fix the text manually for production, and then immediately re-engineer the system prompts so the machine doesn't make the same mistake twice.

Accuracy Verification: Ensure that the engineered prompts extract and maintain profound historical and archaeological accuracy without losing narrative tension.

Data-Driven Iteration: Cross-reference YouTube Studio retention graphs with our script templates to identify where viewers get bored, and systematically update the prompt logic to fix those specific algorithmic leaks.

Requirements:

Proven experience writing advanced, structured prompts for LLMs/AI systems (Claude, ChatGPT, or APIs) to output specific formatting and behavioral constraints.

Exceptional analytical editing skills. You look at a script like a machine and know exactly when a paragraph will cause a viewer to click away.

Solid understanding of YouTube analytics (specifically AVD, retention charts, and viewer drop-off psychology).

Flawless written English and an eagle eye for pacing, conversational flow, and structural narrative tension.

Rate, Type & Career Growth:

Immediate Start: This begins as a project-based paid trial, estimated at 15–20 hours per week to evaluate your prompt mechanics and structural script approach.

Long-Term Trajectory: While this is a part-time gig at the very start, we are looking for a long-term strategic partner. If your prompt engineering and quality checks successfully move our target tracking metrics, this will transition into a permanent, consistent part-time or full-time strategic role as we scale our network.

Compensation: $10.00 – $15.00 USD/hour (negotiable based on your proven track record with YouTube retention analytics and prompt engineering).

How to Apply (Strict Instructions — Failure to follow means instant rejection):

To apply, you must strictly follow these 3 steps:

Start the very first line of your cover letter with the word "STRATEGY".

Provide a link to a short 60-second audio or video recording (using free tools like ---------- or ---------- ) introducing yourself and mentioning one history channel whose pacing you admire.

Answer this specific diagnostic question: If a long-form video starts with a healthy 70% retention in the first 30 seconds, but suddenly drops sharply by 30?tween the 1-minute and 2-minute mark, what did the script structurally fail to do, and what specific instruction would you add to the AI prompt to prevent this from happening?

SKILL REQUIREMENT
Proofreading Youtube Marketing Prompt Engineering
VIEW OTHER JOB POSTS FROM:
SHARE THIS POST
facebook linkedin
  BENCHMARKS  
Loading Time: Base Classes  0.0008
Controller Execution Time ( Jobseekers / Job )  0.0248
Total Execution Time  0.0262
  GET DATA  
No GET data exists
  MEMORY USAGE  
1,510,144 bytes
  POST DATA  
No POST data exists
  URI STRING  
jobseekers/job/YouTube-Prompt-Engineer-Script-Quality-Controller-AI-Biblical-History-Network-1667829
  CLASS/METHOD  
jobseekers/job
  DATABASE:  onlinejobs (Jobseekers:$db)   QUERIES: 13 (0.0146 seconds)  (Hide)
0.0011   SELECT *
                                FROM exrates
                                WHERE rate_name = 'USD-PHP' 
0.0004   SELECT *
FROM `employer_jobs`
WHERE `job_id` = 1667829
 LIMIT 1 
0.0014   SELECT *
FROM `employers`
WHERE `employer_id` = 911321
 LIMIT 1 
0.0032   SELECT COUNT(DISTINCT t.id) as cnt
FROM `t_thread` `t`
INNER JOIN `t_message` `m` ON `t`.`id` = `m`.`thread_id`
INNER JOIN `t_message_employer` `e` ON `m`.`id` = `e`.`message_id`
LEFT JOIN `t_thread_misc` `misc` ON `t`.`id` = `misc`.`thread_id`
WHERE `t`.`job_id` = 1667829
AND `misc`.`id` IS NULL 
0.0014   SELECT e.business_name, e.logo, e.website, e.rebill_date, e.date_added member_date, hits, DATEDIFF('2026-06-20',ej.date_added) duration_days, DATEDIFF('2026-06-20',e.rebill_date) duration_rebill, ej.*, e.deactivate FROM employers e, employer_jobs ej WHERE e.employer_id = ej.employer_id AND
                                   ((e.user_level >= '500' AND ej.date_added <= e.rebill_date)
                                   OR e.employer_id = '' OR (ej.date_approved <> '2000-01-01' and DATEDIFF('2026-06-20',ej.date_added) <= 14 ))
                                   AND e.deactivate != 1 AND ej.deleted = 0 AND job_id = '1667829' 
0.0003   SELECT *
FROM `employer_jobs_skills` `ejs`
LEFT JOIN `skills_categories` `sc` ON `ejs`.`skill_id` = `sc`.`id`
WHERE `job_id` = 1667829 
0.0017   UPDATE employer_jobs SET hit_counts = '***Jun-13-2026=636***Jun-19-2026=1***Jun-20-2026=1' WHERE job_id= '1667829'  
0.0014   UPDATE employer_jobs SET monthly_hits = '***Jun-2026=638' WHERE job_id= '1667829'  
0.0012   SELECT date_sent FROM jobseeker_sent_emails WHERE jobseeker_id = '' AND job_id = '1667829' AND status LIKE 'sent%' ORDER BY id DESC  
0.0004   SELECT *
FROM `employer_jobs_skills` `ejs`
LEFT JOIN `skills_categories` `sc` ON `ejs`.`skill_id` = `sc`.`id`
WHERE `job_id` = 1667829 
0.0012   SELECT COUNT(*) AS `numrows`
FROM `employer_jobs`
WHERE `employer_id` = '911321'
AND `date_added` >= '2022-06-08' 
0.0007   select * from teasers 
0.0003   SELECT * FROM skill_categories WHERE skill_cat_id='' 
  HTTP HEADERS  (Show)
  SESSION DATA  (Show)
  CONFIG VARIABLES  (Show)