Full Stack AI Engineer

Summary:

We're a small, high-output engineering team building AI-powered SaaS products such as RAG systems, agentic workflows, document pipelines, and automation tooling. You'll work directly with the lead engineer on end-to-end features: designing backend APIs, wiring up LLM integrations, and shipping polished frontends. This is not a checkbox role - you'll be expected to think architecturally, own your deliverables, and move fast.

Responsibilities:

• Build and iterate on RAG pipelines — chunking strategies, hybrid retrieval, reranking, evaluation.
• Develop agentic workflows using LangChain / LlamaIndex with tool use and memory.
• Ship FastAPI backends with async task queues (Celery Redis) and clean data models.
• Build Next.js / React frontends that are fast and production-ready.
• Integrate with Anthropic, OpenAI, and Gemini APIs — switching models based on cost/task.
• Deploy on Kubernetes (AKS/EKS) or cloud services with proper observability.
• Participate in architecture discussions and code reviews.

Requirements:

Experience: 5+ Years

Location: Vizag Office

Required Skills:

• Python
• FastAPI
• LangChain or LlamaIndex
• LLM APIs (OpenAI / Anthropic / Gemini)
• React / Next.js
• TypeScript
• PostgreSQL
• Docker
• REST API design

Preferred Skills:

• Celery Redis
• Kubernetes / AKS
• Azure / AWS
• Terraform
• Langfuse / OpenTelemetry
• n8n automations
• Ollama / local LLMs
• Semantic Kernel
• Vector DBs (Pinecone, Qdrant, pgvector)

Benefits:

• Strong ownership - you ship things end-to-end without hand-holding.
• Comfortable switching between backend, frontend, and AI layers in the same day.
• You read docs, debug independently, and propose solutions proactively.
• Experience building something from scratch - a side project, open-source tool, anything real.
• Can communicate technical trade-offs clearly to non-technical stakeholders.
• Comfortable in fast-moving, small-team environments with evolving requirements.