Conversational AI
Content Creation
↗
↗
↗
↗
↗
Q&A Thumbnails
AI-powered Netflix-style thumbnails for health Q&A content using Gemini image generation
Molecule A-Z
Medical illustrations for molecule pages
Conditions
Disease & condition thumbnails
Posts
Health post thumbnails
Video Content Agent
4-pillar AI topic discovery, script generation, multi-language video tracking & social media posting
Localization & Voice
Data & Research
Growth & Revenue
5
Departments
15
AI Agents
6
Pipelines
3
Resource Lanes
Content Creation Head
Thumbnail design, video content strategy & visual asset generation
Responsibilities
- Generate AI thumbnails for all content types (Q&A, molecules, conditions, posts)
- Maintain consistent Netflix-style medical illustration aesthetic
- Process batch generation through Gemini API with concurrency control
- Upload assets to GCS and update image URLs in production DB
- 4-pillar video content strategy: topic discovery, script generation in 13 languages
- Track video production pipeline and social media distribution across platforms
Agents
Q&A Thumbnail Agent
Generates thumbnails for vertex_chat Q&A posts
Molecule Thumbnail Agent
Medical illustrations for molecule_groups table
Condition Thumbnail Agent
Emotion-driven illustrations for diseases table
Posts Thumbnail Agent
Netflix-style thumbnails for health posts
Video Content Agent
4-pillar topic discovery, OpenAI script generation in 13 languages, video tracking & social posting
Targets
100%
Page coverage
10/batch
Items per cycle
<5%
Failure rate
Timeline
Q&A + Molecule thumbnails Done
Core Gemini image generation pipeline
Condition + Posts thumbnails Done
Extended to all content types
Clickbait hook text overlays Done
Engagement-boosting text on thumbnails
Orchestrator batch processing Active
Automated via Nucleus Orchestrator
Video Content Agent Active
4-pillar playbook, AI topics & scripts in 13 languages, social media tracking
Localization & Voice Head
Multi-language translation & voice synthesis
Responsibilities
- Translate disease content from English to 8 Indian & global languages
- Maintain translation quality via dictionary & review workflow
- Voice cloning and TTS across 21 languages (DashScope, Sarvam, IndicF5)
- Manage voice avatars and preset voices in production DB
Agents
Disease Translator Agent
OpenAI-powered translation with dictionary enforcement
Translation Review Agent
Quality review, retranslation & sync workflows
Voice Clone Agent
Multi-provider voice cloning (DashScope, Sarvam, IndicF5)
Targets
8
Translation languages
21
Voice TTS languages
100%
Hindi coverage
Timeline
Hindi translation pipeline Done
OpenAI-based disease translation
Voice cloning (DashScope + Sarvam) Done
Multi-provider voice clone with 10+ languages
IndicF5 Indian language TTS Done
Zero-shot cloning on GCP GPU VM
Multi-language expansion Active
Extending to all 8 target languages
Data & Research Head
Web scraping, data collection & user intelligence
Responsibilities
- Scrape MHRA SPC documents for drug_labels and ukmhra_labels tables
- Discover trending health questions from Google, Reddit, Quora & YouTube
- Build and maintain the Medwiki Intelligence Graph (user profile inference)
- Ensure data pipelines run reliably with retry and error handling
Agents
MHRA Scraper Agent
Playwright browser automation for drug_labels SPC PDFs
UKMHRA Labels Agent
Batch scraping with auto-agent loop
Health Trends Agent
Multi-source trending question discovery
MIG Intelligence Agent
LLM-powered user profile inference with ChromaDB
Targets
100%
UKMHRA SPC coverage
Weekly
Trends cadence
22+
MIG parameters
Timeline
MHRA SPC scraper Done
Playwright-based with GCS upload
UKMHRA labels + auto-agent Done
Automated loop until all rows processed
Health Trends multi-source Done
Google PAA, Reddit, Quora, YouTube, NDTV, TOI
MIG ChromaDB few-shot Active
Improving inference accuracy
Growth & Revenue Head
SEO optimization & revenue analytics
Responsibilities
- Identify low-CTR pages and generate AI-powered SEO recommendations
- Aggregate revenue data from GA4, AdSense & Google Search Console
- Score pages by revenue potential and surface optimization opportunities
- Generate weekly reports and send alerts to Google Chat
Agents
SEO CTR Agent
GSC integration, Gemini analysis, knowledge base learning
Revenue Engine Agent
GA4 + AdSense aggregation, scoring, weekly reports
Targets
+15%
CTR improvement
Weekly
Report generation
Daily
Data fetch cadence
Timeline
Revenue Phase 1 — Data aggregation Done
GA4 + AdSense + GSC collection
Revenue Phase 2 — Intelligence Done
Page scoring, opportunities & AI insights
Revenue Phase 3 — Reporting Done
Weekly reports with Google Chat delivery
SEO CTR learning loop Active
Knowledge base self-learning from recommendations
Operations Head
Pipeline orchestration & infrastructure
Responsibilities
- Coordinate all 6 content pipelines with resource-aware scheduling
- Manage shared resources: browser lock, Gemini semaphore, OpenAI lock
- Round-robin batch processing across browser, Gemini & OpenAI lanes
- Monitor pipeline health, track progress & handle failures gracefully
Agents
Nucleus Orchestrator Agent
Resource-aware round-robin pipeline runner with 3 lanes
Targets
6
Pipelines managed
24/7
Unattended processing
0
Resource conflicts
Timeline
Pipeline definitions & resource lanes Done
Browser, Gemini & OpenAI lane separation
Round-robin orchestration loop Done
Cycle-based batch processing with stop/resume
Auto-scaling & smart scheduling Active
Dynamic batch sizing based on pipeline health
Cross-pipeline dependency graph Planned
DAG-based ordering and dependency resolution