ai engineer – langchain & llm frameworks specialist

Our Client is seeking an experienced Senior AI Engineer with deep expertise in LangChain and modern LLM frameworks. You will lead the development of advanced LLM-driven applications, design complex RAG architectures, and guide the team in best practices for LLM engineering.

Key Responsibilities

LangChain & LLM Development

• Design and develop advanced applications using LangChain, LangGraph, and LangSmith

• Build complex chains, agents, and multi-agent systems

• Implement custom tools, retrievers, and memory systems

• Develop context-aware conversational AI and chatbot applications

• Create workflow orchestration using LangGraph for complex reasoning tasks

• Integrate multiple LLM providers (OpenAI, Anthropic, Cohere, Azure OpenAI)

• Optimize prompt templates and output parsers

RAG & Vector Database Architecture

• Design and implement advanced Retrieval-Augmented Generation (RAG) systems

• Develop custom embedding strategies and chunking techniques

• Work with vector databases such as Pinecone, Weaviate, Chroma, Qdrant, Milvus

• Implement hybrid search (semantic + keyword) and reranking methods

• Build document loaders and text splitters for diverse data sources

• Optimize retrieval quality with metadata filtering and query expansion

• Develop multi-modal RAG systems (text, images, documents)

Advanced LLM Techniques

• Apply advanced prompt engineering: few-shot, chain-of-thought, ReAct

• Build function calling and tool-use patterns

• Develop structured output with Pydantic models

• Implement streaming responses and async pipelines

• Build efficient caching strategies for cost optimization

• Implement fallback mechanisms and robust error handling

• Integrate guardrails and content moderation systems

Production & Monitoring

• Deploy LangChain applications to production using LangServe

• Integrate LangSmith for tracing, debugging, and evaluation

• Implement monitoring and observability for LLM applications

• Build evaluation frameworks for quality assurance

• Optimize latency, cost, and token usage

• Run A/B testing for prompt variations

• Create dashboards to monitor key LLM metrics

Required Qualifications

Education

• Bachelor’s degree in Computer Science, Machine Learning, or related field (required)

• Master’s degree in AI/ML or Computer Science (preferred)

Experience

• 5+ years of software development experience

• 3+ years working with AI/ML and LLM applications

• 2+ years hands-on experience with LangChain (mandatory)

• Proven experience building production-grade LLM applications

• Strong background with RAG architectures and vector databases

Technical Skills – LangChain Ecosystem

 LangChain Core (Required)

• Expert-level knowledge of LangChain (v0.1+)

• Deep understanding of Chains, Agents, Tools, Memory, Callbacks

• Experience with LangChain Expression Language (LCEL)

• Knowledge of LangGraph for complex workflows

• Familiarity with LangSmith for monitoring/debugging

• Experience deploying with LangServe

• Understanding of LangChain Templates

LLM Providers & APIs

• Strong experience with OpenAI API (GPT-4, GPT-3.5, embeddings)

• Knowledge of Anthropic Claude API

• Familiarity with Azure OpenAI Service

• Experience with open-source models (Llama, Mistral, Mixtral)

• Knowledge of Hugging Face Inference API

• Understanding model capabilities, limitations, and pricing models

Vector Databases & Embeddings

• Hands-on experience with Pinecone, Weaviate, or Chroma

• Knowledge of embedding models (OpenAI, Cohere, sentence-transformers)

• Understanding similarity search algorithms

• Experience with metadata filtering and hybrid search

• Knowledge of vector indexing strategies (HNSW, IVF)

Programming & Software Engineering

• Expert-level Python (asyncio, type hints, Pydantic)

• Experience with FastAPI

• Strong understanding of async/await patterns

• Strong software engineering fundamentals

• Experience with Git, CI/CD, testing frameworks

 Cloud & Infrastructure

• Experience with AWS (Lambda, ECS, S3, DynamoDB) or Azure

• Docker containerization

• Kubernetes (preferred)

• Serverless architectures

• API Gateway and load balancing

Core Capabilities

• Deep understanding of LLM capabilities and limitations

• Expertise in prompt engineering and optimization

• Ability to design scalable LLM architectures

• Strong understanding of cost optimization for LLM workloads

• Knowledge of AI safety and responsible AI practices

• Excellent problem-solving and communication skills

Preferred Qualifications

• Contributions to LangChain or other open-source LLM projects

• Experience with LlamaIndex, Haystack, or similar frameworks

• Knowledge of fine-tuning (LoRA, QLoRA)

• Experience with multi-agent systems and AutoGPT patterns

• Knowledge of semantic caching and query optimization

• Experience with document intelligence and OCR

• Understanding transformer architectures

• Experience with evaluation frameworks (RAGAS, TruLens)

• Experience with streaming and real-time LLM processing

• Blog posts, tutorials, or presentations on LangChain

• Cloud certifications

What They Offer

• Highly competitive salary and equity package

• Full health, dental, and vision insurance

• Generous budget for professional development and certifications

• Flexible working arrangements (remote/hybrid options)

• Access to cutting-edge AI infrastructure and LLM APIs

• Collaborative environment with world-class AI professionals

• Participation in conferences and AI community events

• Opportunities for technical leadership and mentorship

• Influence over the company’s AI strategy

 Success Metrics

Success in this role will be measured by:

• Quality and performance of LangChain applications

• Reduction in latency and improved cost efficiency

• Accuracy and relevance of RAG systems

• Adoption of best practices across the team

• Speed of delivery for new AI features

• User satisfaction with AI-powered features

• Technical leadership and mentorship impact

• Contribution to documentation and knowledge base

Technologies You Will Work With

LLM Frameworks

• LangChain, LangGraph, LangSmith, LangServe

• OpenAI API, Anthropic Claude API

• Hugging Face Transformers

Vector Databases

• Pinecone, Weaviate, Chroma, Qdrant

• PostgreSQL with pgvector

Development Stack

• Python 3.11+, FastAPI, Pydantic

• Docker, Kubernetes

• AWS/Azure cloud services

• Git, GitHub Actions

Monitoring & Observability

• LangSmith

• Prometheus, Grafana

• DataDog, New Relic