Full Stack Automation Engineer

WHO WE ARE

We've built an AI-native internal platform that powers every aspect of our Amazon brand management business. AI isn't a feature — it's the backbone.

- LLMs classify and respond to inbound communications

- AI generates pre-call intelligence briefs from raw enrichment data

- A RAG system feeds context into every generation pipeline

- An AI checkpoint system audits all generated content against quality gates

The platform is already live and scaling fast:
• 17+ background services
• 130+ frontend pages
• 214 backend services
• 184 database tables
• Dozens of autonomous AI pipelines
OUR CORE VALUES
Be A Moral Human

Serve a higher purpose. Everything we build and every decision we make is grounded in doing the right thing.
Improve 1% Daily

Strive for 1% better every day. Consistent, compounding improvement is how we build world-class systems and teams.
Extreme Ownership

Own your actions. No excuses, no hand-offs as a crutch. If it's in your world, it's your responsibility.
Solve Problems, Don't Create Them

Every challenge has a solution. Don't slow down the team by creating new problems — come with answers, not blockers.
Make an Impact

Focus on meaningful results. We measure success by the difference we make, not the hours we log.
Fail Fast & Fail Forward

Don't be afraid to fail. Take that failure, learn from it, and move forward stronger. Every failure is a lesson.
WHAT YOU'LL BUILD & SCALE
AI Communication Pipelines

- Classify inbound messages by category, intent, urgency, and tone

- Generate contextual responses using enrichment data

- Implement and tune human approval gates

AI-Powered Sales Intelligence

- Transform raw enrichment data into structured pre-call briefs

- Generate backgrounds, pain hypotheses, talking points, and rapport hooks

RAG System

- Maintain and improve the vector database with embeddings

- Implement markdown-aware chunking strategies

- Build async ingestion workers and semantic search APIs

Trend Intelligence Engine

- Process RSS feeds, social media, video platforms, and search trends

- Generate reports, forecasts, and content drafts

• Run autonomously on scheduled jobs
Content Quality Pipeline

- Extend the multi-agent system (outline → audit → generate)

- Maintain binary quality gates (PASS/FAIL with citations)

- Support multiple content formats across the pipeline

Automated Lead Qualification

- Enrich leads with product data and market insights

- Build AI scoring and qualification grading systems

• Generate automated audit reports
AI Executive Assistant

- Build and maintain Slack-integrated operations

• Automate scheduling workflows

- Triage and respond to email autonomously

Requirements
DAY-TO-DAY RESPONSIBILITIES

- Build and improve AI pipelines for client performance insights

- Improve RAG retrieval quality (re-ranking, chunking, hybrid search)

- Add tool use / function calling for real-time data in LLM pipelines

- Debug classification errors and improve model accuracy

- Optimize LLM costs, latency, and performance

- Build dashboards for AI metrics and usage monitoring

- Add observability and tracing to AI pipelines

- Expand content quality systems to new formats and use cases

TECH STACK

Core: TypeScript · Node.js · React / Next.js · n8n · PostgreSQL · CI/CD · Claude Code

Nice to have: AWS Lambda · Terraform · Docker · Amazon SP-API · Slack Bots · Playwright
QUALIFICATIONS

Required:

- Production LLM experience — Claude or OpenAI deployed in real, live systems

- RAG system experience — embeddings, retrieval, chunking, and context handling

• 3+ years TypeScript / Node.js

- Strong React skills (component architecture, state management, performance)

- PostgreSQL — queries, migrations, indexing, query optimisation

- API integrations — REST, OAuth, webhooks

- Linux server experience — SSH, log analysis, debugging, deployments

Strong Pluses:

- Multi-agent LLM systems and orchestration

- Anthropic Claude exp