We are building Ava, an in-house AI platform that transforms how our agency operates. Ava combines our proprietary data warehouse (SQL Server with daily Amazon metrics across all clients) with AI agents that automate analysis, generate reports, enforce compliance, and provide strategic recommendations. Our goal: scale from 14 to 40 clients while reducing headcount per client – only possible through intelligent automation.
This is not a typical development role. You will be building the AI infrastructure that powers a next-generation Amazon agency.
Core Responsibilities
- Finish and deploy our Next.js analytics dashboard (Polaris) – client-facing pages for Customers, Subscribe & Save, and Lifetime Value, powered by SQL Server
- Set up a vector database and RAG pipeline to unify our unstructured knowledge (SOPs, analysis frameworks, compliance documents, meeting transcripts) so AI agents can pull the right context instantly
- Deploy the platform to production so clients can access their dashboards
- Build production AI agents from our library of 35+ detailed agent specifications – these handle listing optimization, competitive analysis, keyword strategy, compliance checking, and client reporting
- Implement agent-to-agent handoff systems where one agent’s output feeds into the next (e.g., Semantic Strategist → Listing Optimizer → Compliance Checker)
- Build automated alerting – Buy Box monitoring, inventory warnings, performance anomalies – delivered via Slack or email
- Automated client reporting (daily/weekly performance summaries generated from SQL data)
- Client onboarding automation (call transcripts → task generation)
- Monthly Business Review deck generation
- Model fine-tuning and optimization as agent usage scales
- 3–5 years of professional software development experience
- Python – you’ll build AI pipelines, data processing scripts, and backend services
- TypeScript / Next.js / React – our dashboard platform runs on Next.js with Server Components
- SQL – comfortable writing complex queries against SQL Server (our data warehouse has 50+ tables)
- LLM API integration – production experience with Claude API, OpenAI API, or similar. Not just calling an API – building reliable systems with error handling, token management, prompt versioning
- Vector databases and RAG – experience with Pinecone, Weaviate, ChromaDB, pgvector, or similar. Understanding of embeddings, chunking strategies, and retrieval quality
- Agent architectures – building multi-step AI workflows with tool use, validation gates, and state management
- Experience with Amazon SP-API or Amazon Seller Central data
- ML fundamentals (fine-tuning, classification, embeddings beyond RAG)
- Cloud deployment (Vercel, AWS, or similar)
- Experience in an agency or e-commerce environment
- Familiarity with MCP (Model Context Protocol) servers
- RPA tool specialists (UiPath, Blue Prism, Automation Anywhere)
- Pure data scientists who don’t build production systems
- Frontend-only developers with no AI/backend experience
- People who need detailed tickets – you’ll get requirements in plain English and figure out the technical approach
- Location: Remote. Preference for Indonesia, Philippines, or Latin America
- Hours: Full-time (40 hrs/week) – some overlap with US Eastern Time required for team syncs
- Compensation: Competitive, based on experience. Please include your rate expectations in your application
- Tools: Claude Code, VS Code, Git, SQL Server Management Studio, Slack
Submit the following:
- Resume or portfolio highlighting AI/ML and full-stack projects you’ve built
- Brief note (2–3 paragraphs) on why this role interests you and what you’d bring to it
Kindly submit your application at [email protected], using the subject line: Applicant – AI Platform Engineer.