Company Description
BolsterBiz is an offshoring and outsourcing company based in New Delhi, serving 100+ US clients across customer support, data analytics, digital marketing, and IT staffing. We've completed over 2,000 projects and are now building our first internal product an AI-powered customer support platform. We're looking for the engineer who will take it from idea to live product.
The Role
Responsibilities
You'll be our first engineering hire, responsible for building the MVP of an AI-powered customer support product with two core modules:
- Help Article Generator A tool that ingests a client's product documentation and a short founder survey, then uses AI to auto-generate a structured help center / knowledge base.
- AI L1 Support Agent A chatbot that answers end-customer queries using the generated knowledge base, and intelligently escalates unresolved tickets to a human agent.
This is a hands-on, full-stack building role. You'll own the product technically from the RAG pipeline to the chat interface to the escalation logic. There's no existing codebase; you're starting from zero.
What You'll Build (First 90 Days)
- A RAG pipeline that processes product documentation and founder survey responses into a searchable knowledge base using vector embeddings
- An AI-powered chat agent (built on Claude / OpenAI) that retrieves relevant context from the knowledge base and responds to customer queries in natural language
- Confidence-based escalation logic that detects when the AI cannot answer reliably and routes the conversation to a human agent with full conversation context and summary attached
- A lightweight admin dashboard where clients can review, edit, and approve auto-generated help articles, view conversation logs, and track agent performance metrics
- A simple, embeddable web chat widget that clients can drop into their website or app
What You'll Need
Required (non-negotiable):
- 35 years of professional software engineering experience
- Hands-on experience building with LLM APIs (Claude, OpenAI, or similar) side projects and personal builds count, but you must have shipped something functional.
- Working knowledge of RAG pipelines: text chunking, embeddings, vector databases (Pinecone, Qdrant, Weaviate, or pgvector), and retrieval strategies
- Backend proficiency in Python (FastAPI preferred) or Node.js/TypeScript
- Comfortable building frontend interfaces in React (admin dashboards, chat UIs doesn't need to be pixel-perfect, but must be functional and clean)
- Solid experience with PostgreSQL or a similar relational database
- Ability to work independently with minimal direction you'll be the only engineer for the first few months
Nice-to-have (will Accelerate Your Impact)
- Experience with prompt engineering and conversation design (fallback logic, hallucination guardrails, tone calibration)
- Familiarity with orchestration frameworks like LangChain or LlamaIndex
- Prior experience at an early-stage startup or as a solo / founding engineer
- Exposure to customer support or ticketing APIs (Zendesk, Intercom, Freshdesk)
- Experience setting up basic CI/CD pipelines and containerized deployments (Docker, AWS/GCP)
What We're NOT Looking For
- System architects who want to design for 6 months before shipping a line of code
- Engineers who need a fully detailed PRD handed to them before they can start building
- People optimizing for a management title this is a heads-down building role first
What You Get
- Ownership from day one you choose the tech stack, you make the architecture calls, you ship the product
- Direct access to the founders no layers between you and business context; you'll shape the product, not just code it
- Clear growth path as the product scales post-MVP, you'll hire your team and grow into a technical leadership role
