Overview

Company Description

Responsibilities

• Design and implement production-grade LLM applications
• Build and scale Retrieval Augmented Generation (RAG) systems
• Implement tool / function calling, structured outputs, and guardrails
• Develop hybrid retrieval systems (BM25 + vector search)
• Build cross-LLM architectures supporting multiple model providers
• Optimize AI systems for cost, latency, and accuracy
• Design and implement agent-based architectures
• Understand and apply:
• Agent planning vs reactive agents
• Tool execution and retry strategies
• State management in agents
• Multi-agent coordination
• Failure detection and recovery mechanisms
• Use frameworks such as LangChain and LangGraph effectively
• Work with vector databases and scale them efficiently
• Build and maintain data pipelines for AI systems
• Handle large datasets using Pandas / Polars
• Implement data cleaning and preprocessing workflows
• Apply SQL fundamentals for data querying and analysis
• Design inference pipelines for batch and streaming workloads
• Apply prompt engineering techniques in depth
• Understand GPU fundamentals (memory, compute, inference constraints)
• Work with model fine-tuning concepts and evaluate when fine-tuning is needed
• Implement AI cost optimization strategies (model selection, token optimization, caching, architecture decisions)
• Translate business requirements into technical AI solutions
• Clearly communicate trade-offs between cost, accuracy, latency, and complexity
• Know when AI is not the right solution and propose alternatives
• Collaborate with Product Managers, Engineers, and Clients
• Write clear and maintainable technical documentation

Qualifications

• Python (advanced)
• LangChain, LangGraph
• NLP systems and production AI pipelines
• RAG architectures
• Vector databases (design, scaling, optimization)
• Hybrid search (BM25 + vectors)
• Agentic AI design
• Prompt engineering (advanced)
• GPU basics
• Model fine-tuning concepts
• SQL
• Pandas / Polars
• Strong problem-solving mindset
• Clear verbal and written communication
• Ability to explain complex systems to non-technical stakeholders
• Ownership and accountability
• Exposure to MLOps / LLMOps practices
• Prior client-facing experience
• eoffs (cost vs accuracy)
• Writes documentation
• Can work with PMs & clients