Context Engineer
The evolution of Prompt Engineer — architects the complete information environment for AI interactions, not just individual prompts. Selects, structures, and delivers the right documents and data into the LLM context window so the model reliably solves complex tasks without hallucination. Coined by Shopify CEO and Andrej Karpathy in mid-2025.
- Salary Range
- $130K–$220K+
- Hiring Companies
- OpenAI · Shopify · Google
- Hot Regions
- USAU.K.SingaporeEuropean Union
- Core Skills
- LLM · RAG · System Design
Learn this at JR Academy
Real bootcamps currently running on jiangren.com.au — click through for the full outline and enrollment
Job Description Analysis
Core Responsibilities
- Design and maintain the full context architecture for LLM applications, including system prompts, retrieval layers, memory, and tool context
- Optimize context-window usage so models receive high-signal information without wasting tokens
- Build dynamic context-assembly pipelines that select and order the right snippets for each request
- Define context-quality metrics such as relevance, freshness, and completeness, then monitor them continuously
- Work with RAG and agent engineers to keep context consistent across multi-step workflows
Requirements
Tap one answer per requirement: I have it, not yet, or not sure. Your first gap view appears below straight away. No sign-in needed; until you save, your answers stay on this page and clear when you refresh.
- Deeply understand the Transformer architecture and attention mechanism, and know how context is processed inside the model
- Proficient in at least one LLM orchestration framework (LangChain, LlamaIndex, Semantic Kernel)
- 3+ years of backend engineering experience, familiar with distributed systems and caching strategies
- Master vector databases (Pinecone, Weaviate, Qdrant) and document processing pipelines
- Excellent system design capabilities, able to make trade-offs between latency, cost and quality
Your first gap view
Your own answersAnswer one requirement and this panel lists what is missing and which transition step on this page covers it.
Want it more accurate?
This view comes only from your own answers. Add material and the evidence level goes up:
- Upload your resume for the AI resume review
Evidence becomes: from your resume
- Match your resume against a real job ad with Job Hunter
Evidence becomes: resume × job ad
- Take the career impact quiz
Evidence becomes: measured
Hands-on work and projects happen in the CareerOS desktop app.About CareerOS desktop
Nice to Have
- Experience in large-scale RAG system production
- Learn about GraphRAG and knowledge graph technology
- Experience in Agent orchestration and multi-round dialogue state management
A Day in the Life
- Morning: Review context-quality metrics such as hit rate, token waste, and hallucination rate
- Morning: Work with the agent team on context-delivery patterns for multi-step tool use
- Afternoon: Rebuild chunking or retrieval logic to improve recall and reduce irrelevant context
- Afternoon: Decide what belongs in the system prompt, what should come from retrieval, and what should be fetched by tools
- Evening: Study how long-context models change architecture decisions, not just benchmark numbers
Career Transition Guide
Suitable Backgrounds
- Backend/Platform Engineer: Experienced in system design and distributed architecture, quick to get startedSee this career today: Platform Engineer
- Search Engine Engineer: Understand information retrieval and sorting, directly transfer core skills
- Data Engineer: Familiar with data pipelines and ETL, document processing pipelines are essentially similarSee this career today: Data Engineer
- Prompt Engineer: Upgrade to full-link context management based on Prompt
Transition Path
- Step 1: Learn deeply about Transformer architecture and attention mechanism principles, and understand the nature of context windows
- Step 2: Use LangChain to build a complete RAG application and experience the entire process of document slicing → Embedding → retrieval → generation
- Step 3: Learn the core technologies of context engineering: dynamic prompt assembly, context compression, and memory management
- Step 4: Practice context sharing and delivery strategies for multi-agent scenarios in open source projects
- Step 5: Write your experience into a technical blog and submit for the position of Context Engineer or Senior AI Engineer
Learning Roadmap
Phase 1: Theory (January-February)
Phase 2: Engineering Practice (February-April)
Phase 3: Architecture Design (April-June)
Common Mistakes
Think Context Engineering is just another name for Prompt Engineering
Context Engineering is a system-level architecture design, covering complete engineering links such as retrieval, caching, and orchestration.
Squeeze all the information into the context window, the more the better
Quality of context is far more important than quantity - irrelevant information dilutes attention and reduces output quality
Only focus on the context of a single round of dialogue, ignoring multi-round and multi-Agent scenarios
80% of the challenges in real business lie in multi-turn dialogue and cross-Agent context consistency
Recommended Resources
Ready?
Start learning the core skills for Context Engineer
Start Learning →