JOB-02 // APPLIED AI & RAGAI & Machine LearningRemote — Worldwide (EU / PK Overlap) (REMOTE)FULL TIMEMid-Senior (3+ Years)

Applied AI, RAG & Agentic LLM Systems Engineer

Design and deploy hybrid dense + BM25 Retrieval-Augmented Generation (RAG) pipelines, pgvector/Qdrant knowledge indices, and deterministic tool-calling AI agents with strict citation verification.

Compensation Band: USD 28,000 – 54,000 per year

Role Overview at NEURALXCIPHER (PRIVATE) LIMITED

NEURALXCIPHER (PRIVATE) LIMITED builds production AI systems where hallucination tolerance is zero — across FinTech, digital healthcare, and enterprise knowledge bases. You will engineer end-to-end RAG ingestion pipelines, semantic chunking strategies, cross-encoder reranking services, and low-latency SSE streaming APIs in Python (FastAPI) and TypeScript (NestJS).

What You Will Architect & Ship

Architect hybrid retrieval pipelines combining dense vector embeddings (pgvector / Qdrant / Pinecone) with sparse lexical BM25 search and Reciprocal Rank Fusion (RRF).
Implement deterministic JSON-schema tool-calling agents and multi-step orchestration using OpenAI, Anthropic Claude, and Google Gemini APIs.
Build automated RAG evaluation harnesses measuring context recall, groundedness, citation accuracy, and TTFT (Time-To-First-Token) latency.
Enforce PII redaction, tenant-isolated vector namespaces, and cryptographic audit logging for enterprise compliance.
Collaborate with mobile and web teams to stream structured AI citations and interactive UI cards over Server-Sent Events (SSE).

Required Technical Qualifications

3+ years of software engineering experience with 1.5+ years building production LLM, RAG, or NLP systems beyond prototype notebooks.
Proficiency in Python (FastAPI, Pydantic, asyncio) and TypeScript/Node.js.
Hands-on experience with vector databases (PostgreSQL pgvector, Qdrant, or Pinecone) and embedding/reranking models.
Deep understanding of token economics, semantic caching, prompt injection defenses, and structured outputs.

Bonus & Nice-to-Have Experience

Experience with on-device inference (ONNX Runtime, CoreML, TensorFlow Lite) for mobile apps.
Familiarity with Supabase, Docker, and cloud deployment on AWS/GCP.
Contributions to AI/ML research or technical writing.

Compensation, Perks & Engineering Environment

100% Remote work with flexible async engineering culture.
Competitive USD-pegged compensation and annual learning budget.
Access to frontier LLM APIs, GPU compute credits, and modern observability tooling.
Co-author technical research articles and open-source AI benchmarks with our engineering leadership.

// CORE TECHNOLOGY STACK FOR THIS ROLE

PythonFastAPITypeScriptpgvectorQdrantOpenAIClaudeGeminiSupabase