/
JOB DESCRIPTIONS
What You'll Actually Do
You won't be writing CRUD APIs or building login pages. Here's what a typical week looks like:
Build a conversational AI system that processes payment data and answers business questions in natural language
Design a RAG pipeline that ingests 500+ documents and serves accurate answers with source citations
Integrate Claude/OpenAI APIs into a Next.js app with streaming responses, error handling, and cost optimization
Write and iterate on prompts until they produce structured, reliable outputs across edge cases
Set up evaluation pipelines to measure AI output quality before shipping
Debug why the embeddings retrieval is returning irrelevant chunks and fix the chunking strategy
Deploy an AI automation workflow that replaces 10 hours of manual work for a client
Must-Have Skills (Non-Negotiable)
AI/LLM (this is the core of the role):
You have built and deployed at least ONE AI-powered feature or product (not a tutorial, not a Kaggle notebook - something with real users)
Working experience with LLM APIs (Claude, OpenAI, or similar)
Understanding of RAG - you know how to chunk documents, generate embeddings, store them in a vector DB, and retrieve relevant context
Prompt engineering for structured outputs - not just chatting with ChatGPT, but designing prompts that return reliable JSON/structured data
Familiarity with at least one: LangChain, LlamaIndex, Vercel AI SDK, or similar orchestration framework
Engineering (you must be a strong developer first):
3–5 years of professional software development experience
Strong JavaScript/TypeScript - this is our primary language
React or Next.js - you've built complex frontends, not just todo apps
Node.js backend development
API design and third-party integrations
SQL or NoSQL database experience
Git, CI/CD, deployment workflows
Good-to-Have Skills
Experience with vector databases (Pinecone, ChromaDB, Weaviate, Qdrant)
Fine-tuning or training ML models
AI agent frameworks (CrewAI, AutoGen, LangGraph)
Supabase, Firebase, or similar BaaS platforms
Experience with payment systems or fintech
Python (secondary to JS/TS but useful for AI tooling)
What We're Looking For (Mindset)
You use AI daily as a developer. Cursor, Copilot, Claude, ChatGPT - AI is already part of how you write code, debug, and think. If you're still coding everything from scratch without AI assistance, you're behind.
You build, not just learn. We don't care about certifications or courses. Show us something you've shipped.
You think in systems. You don't just complete tickets — you understand how pieces connect and you design for scale.
You move fast. We're a small team. You'll ship features in days, not sprints.
You take ownership. No hand-holding. You pick up problems and solve them end-to-end.
How We Hire (Our Process)
Apply (5 min) Fill out the form on this page. Tell us who you are and why Dapton.ai.
Screening (15–20 min) We send you a short assessment. Our AI reviews every response.
Assignment (24-hour window) A real-world task that mirrors what you'd actually do here. No trick questions.
Founder Chat (45 min) A two-way conversation. We learn about you, you learn about us.
Decision within 2 days. We don't ghost.
Perks
Fully remote - work from anywhere in India
Work directly with founders - no corporate bureaucracy
Build real AI products used by real businesses
Early-stage team - massive learning and growth opportunity
Exposure to payment processing, SaaS, and AI industries
Competitive salary
/
APPLICATION