LLM Engineer

Job Title: LLM Engineer Location: Noida Experience: 4+ years Employment Type: Ful-Time Role Overview We are seeking a LLM Engineer to design, optimize, and deploy Large Language Model (LLM) powered systems for real-world applications. This role focuses on prompt engineering, retrieval-augmented generation (RAG), structured outputs, evaluation frameworks, and production-grade LLM integrations on cloud platforms.

Required Skills & Qualifications • Strong hands-on experience with Large Language Models and generative AI systems. • Expertise in prompt engineering and RAG architectures. • Proficiency in Python and backend API development. • Experience with vector databases (FAISS, Pinecone, OpenSearch, or similar). • Cloud experience with AWS, Azure, or GCP.

Good to Have Skills • Experience with agentic or multi-step LLM workflows. • Exposure to voice-based AI pipelines (speech-to-text or text-to-speech). • FinTech, lending, or document intelligence domain experience. • Experience integrating LLMs with CRM platforms such as Salesforce. • Familiarity with LLM evaluation frameworks and observability tools

Key Responsibilities – LLM Engineering • Design and optimize prompts, system instructions, and templates for consistent and accurate LLM behavior. • Build and maintain retrieval-augmented generation (RAG) pipelines for document intelligence and knowledge grounding. • Implement structured outputs using JSON schemas, function calling, and validators. • Reduce hallucinations through grounding, verification, and post-processing techniques. Model Evaluation & Optimization • Define evaluation metrics for accuracy, faithfulness, latency, and cost. • Benchmark different LLMs and embedding models for specific use cases. • Optimize token usage, response latency, and inference cost in production environments. Cloud & Production Deployment • Deploy LLM-powered services on AWS, Azure, or GCP. • Integrate with managed LLM platforms such as AWS Bedrock or Azure OpenAI. • Build scalable APIs, background workers, and batch processing pipelines. • Implement logging, monitoring, and alerting for LLM applications. Integration & Data Systems • Integrate LLM systems with databases, vector stores, and internal APIs. • Design data ingestion, chunking, embedding, and retrieval strategies. • Work closely with product, data, and platform teams to embed LLM capabilities into business workflows. Governance, Security & Quality • Implement guardrails, prompt versioning, and auditability for LLM outputs. • Ensure compliance with security, privacy, and enterprise AI standards. • Document LLM behavior, limitations, and known failure modes.

Avants AI partners with companies to vet candidates with AI-powered interviews and match scoring. Every application in our pipeline is already screened before a human reads it. We consider all qualified applicants without regard to legally protected characteristics.