Artificial Intelligence Engineer

Job Title: Artificial Intelligence Engineer (LLM & Agentic AI)Company: NEPAWorksDepartment: AI & Data InnovationPosition Type: Full-TimeShift Timing: Night ShiftWork Mode: On-site (Kathmandu)Company DescriptionNEPAWorks is a global shared services leader, specializing in integrating core business functions into a unified and scalable ecosystem. Through its Global Shared Service Ecosystem, which includes innovative solutions such as FinanceWorks, TechWorks, DataWorks, BrandWorks, MarketWorks, ClientWorks, and PeopleWorks, NEPAWorks enables businesses in North America, South America, and Europe to optimize operations and achieve growth. With a focus on agility and innovation, NEPAWorks provides AI-driven data analytics, intelligent technology systems, and transformative branding and marketing services. The organization is dedicated to empowering businesses to excel and maintain a competitive edge, guided by its global standards and commitment to exceptional service.Role DescriptionThis is a full-time Mid-Level Artificial Intelligence Engineer position focused on designing, building, and deploying LLM-based systems and Agentic AI solutions.The role involves developing:Intelligent multi-agent workflowsLLM-powered automation pipelinesRAG (Retrieval-Augmented Generation) applicationsKnowledge systems and vector-based searchTool-using agents capable of reasoning, planning, and task executionThe engineer will work closely with the DataWorks and TechWorks teams to convert business requirements into autonomous, scalable, and production-grade AI systems used across NEPAWorks’ global service ecosystem.This role is ideal for someone with 2–4 years of experience in modern AI engineering, strong NLP fundamentals, and hands-on experience in LLM orchestration frameworks (LangChain, LlamaIndex, AutoGen, CrewAI, etc.).Key Responsibilities1. LLM Model EngineeringBuild and fine-tune workflows using GPT, Claude, Llama, or open-source LLMs.Develop structured prompting strategies, chain-of-thought flows, output validation, and role-based reasoning agents.Implement advanced techniques such as:Function calling / tool callingLLM planning and task decompositionGuardrails for safety, reliability, and consistency.2. Agentic AI & Multi-Agent SystemsDesign and develop AI Agents using frameworks like AutoGen, CrewAI, LangChain Agents, or custom architectures.Build agents capable of tool usage (search tools, internal APIs, databases, document retrieval, etc.).Implement memory systems, context management, and autonomous task loops.Develop multi-agent collaboration workflows (planner-executor, supervisor-worker, or role-based agents).3. Retrieval-Augmented Generation (RAG)Build RAG pipelines with structured document loaders, chunking logic, and embedding strategies.Work with vector databases such as FAISS, Pinecone, Chroma, or Weaviate.Optimize retrieval quality, ranking metrics, and context packing strategies.4. Backend Integration & DeploymentDevelop backend services (FastAPI/Flask) to integrate LLM and agent workflows into production systems.Deploy AI services to cloud environments (AWS/GCP/Azure) or internal servers.Implement model versioning, endpoint monitoring, and performance analytics.5. Research, Evaluation & OptimizationContinuously evaluate new LLM techniques, agentic frameworks, and improvements in prompt engineering.Analyze system performance, reduce hallucinations, and optimize reasoning accuracy.Produce documentation, architecture diagrams, and implementation cycles.Required Qualifications & SkillsTechnical Skills (Must Have)2–4 years of hands-on experience in AI/ML engineering with a focus on LLMs and NLP.Strong proficiency in Python and modern AI libraries (Transformers, LangChain, LlamaIndex).Experience developing AI Agents with frameworks like: AutoGen, CrewAI, LangChain AgentsCustom autonomous agent systemsDeep understanding of embeddings, text chunking, semantic search, and retrieval pipelines.Strong experience with vector stores such as FAISS, Pinecone, Weaviate, or Chroma.Experience building backend integrations using FastAPI or Flask.Familiarity with cloud platforms (AWS/GCP/Azure) for deploying AI services.Solid understanding of NLP fundamentals: tokenization, semantic similarity, transformers.Preferred SkillsExperience with orchestration tools (Airflow, Prefect) for pipeline automation.Familiarity with evaluation frameworks for LLMs (LLM-as-a-Judge, model eval harness).Knowledge of distributed retrieval, caching layers, or hybrid RAG strategies.Experience with prompt optimization and response validation frameworks (Guardrails, Pydantic-based schemas).Education:Bachelor’s or Master’s degree in Computer Science, Artificial Intelligence, Data Science, Engineering, or related field.Strong portfolio, GitHub contributions, or prior project experience with LLM/Agents is highly valued.Why Join NEPAWorks ?Work on real, enterprise-grade Agentic AI and LLM automation systems.Contribute to global-scale AI initiatives across multiple business ecosystems.Highly collaborative engineering environment with ample innovation opportunities.Rapid growth potential into Senior AI Engineer, Agent Architect, or AI Solutions Lead roles.