Recently, Tobi Lütke, CEO of Shopify, shared an interesting idea on Twitter. He said he prefers the term “context engineering” over the commonly used “prompt engineering” when working with AI tools like ChatGPT. Andrej Karpathy, a prominent AI researcher, supported this view and added valuable insights. Infact we built openalgo and many other tools using LLMs and Agentic Coding tools via Context Engineering rather than traditional prompt engineering
So, what exactly is context engineering, and why does it matter to traders?
The Problem with “Prompt Engineering”
If you’ve used AI tools, you’re familiar with prompts—short commands you give the AI, like: “Write a short summary of today’s market.” But prompts often feel like tossing random tasks to a smart but unaware assistant. You never know if the AI has enough information to understand exactly what you want.
For traders, who rely on precise and accurate insights, generic prompts can produce weak or irrelevant responses.
Enter Context Engineering
Context engineering is about carefully preparing the information you feed into an AI so it can deliver better results. Instead of quick, generic instructions, you provide clear context—relevant background information, examples, and your specific needs.
Imagine asking the AI to build you a trading strategy. A simple prompt might say: “Give me a profitable trading strategy for the stock market.” The AI may respond vaguely.
But with context engineering, your request could look like this:
“I trade Indian stocks using technical indicators. Provide a simple moving-average crossover strategy. I prefer short-term trades and currently focus on large-cap stocks like Reliance or Infosys. Include specific entry and exit rules and explain it clearly.”
Now, the AI knows exactly what you need and can deliver tailored, practical advice.
Why Context Matters More Than Ever
Traders can’t afford misunderstandings or ambiguity. You need actionable insights, clear explanations, and precise strategies. Context engineering ensures the AI has enough information to provide these accurately.
Consider an example:
- Prompt Engineering:
“Tell me how the Indian market performed today.” - Context Engineering:
“Summarize today’s market movements focusing on Nifty and Bank Nifty. Highlight the biggest gainers and losers, major sectors, and any key economic events influencing the market today.”
The second request provides the AI with clear context, leading to a more relevant and useful answer.
Practical Ways Traders Can Use Context Engineering
Here’s how traders can practically use context engineering in daily activities:
- Market Analysis:
Provide details about your watchlist, market conditions, and preferences. The AI then offers customized daily summaries instead of generic market updates. - Trading Strategy Development:
Specify your risk tolerance, favorite technical indicators, asset type, and trading style. This helps AI generate more suitable trading strategies and setups. - Performance Reviews:
Feed the AI information about recent trades and ask it to identify strengths and weaknesses in your strategy or execution. - News Summaries:
Provide specific companies, sectors, or topics you’re interested in. The AI can then filter and summarize news specifically relevant to your trading needs.
Moving Beyond Simple Prompts
Context engineering isn’t just a new buzzword. It’s about clearly communicating your exact needs to AI to receive better, more accurate responses. It transforms a smart but general-purpose assistant into a specialized trading partner who understands your requirements.
As traders increasingly rely on AI, mastering context engineering can set you apart, ensuring you extract maximum value from these powerful tools.
Start today by providing clearer context, not just prompts. Your trading results might thank you for it.
