The Foundation: Role, Context, Task
The quality of any AI-generated output depends entirely on the quality of the prompt. Vague instructions lead to generic, unhelpful results. The most effective prompts are built on a simple but powerful structure: assigning a role to the AI, providing
clear context, and defining a specific task. For example, instead of asking an AI to "write a job description," a strong prompt would begin with, "Act as an expert tech recruiter for the Indian market." This immediately frames the AI's response. From there, adding context about the company, the specific role (e.g., Senior Backend Engineer), and the required tech stack (e.g., Java, AWS, Kafka) guides the AI to produce a far more relevant and targeted output. The final layer is a clearly defined task, such as outlining key responsibilities or generating interview questions.
Crafting Compelling Job Descriptions
A generic job description attracts generic talent. To stand out in the crowded Indian tech market, job posts must be specific and compelling. AI can draft these in seconds, provided you give it the right material. The key is to provide rich context about the company's mission, the challenges the role will tackle, and the team's culture. Avoid asking for a generic list of responsibilities. Instead, focus the AI on the impact the role will have. Example Prompt: "Act as a talent branding expert for a Series B fintech startup in Bengaluru. Write a job description for a 'Senior Frontend Engineer' that focuses on our mission to democratize investing for young Indians. Emphasize growth opportunities and our collaborative, remote-first culture. The tech stack is React, TypeScript, and GraphQL. Avoid corporate jargon and clichés like 'rockstar developer'."
Sourcing and Candidate Discovery
Finding passive candidates—top engineers who aren't actively looking for a job—is a primary challenge. AI can build sophisticated search queries that unearth talent on platforms like LinkedIn, GitHub, and Stack Overflow. Instead of manually creating complex Boolean search strings, you can have the AI do it for you, including variations to capture a wider range of profiles. These prompts can also specify negative keywords to exclude candidates from certain industries or company types, refining the search significantly. Example Prompt: "You are a technical sourcer specializing in high-growth startups in India. Generate an advanced Boolean search string for LinkedIn Recruiter to find 'Software Development Engineer II' candidates in Hyderabad, Pune, or remote. They must have experience with Python and microservices. Exclude anyone currently working at large IT consulting firms."
Screening Resumes at Scale
Manually reviewing hundreds of resumes is a significant time drain. AI can automate this initial screening by comparing a candidate's resume directly against the job description. This is not about automatically rejecting people, but about creating a first-pass filter to highlight the strongest matches for human review. A well-crafted prompt can ask the AI to extract specific skills, identify gaps, and even summarize a candidate's quantifiable achievements. Example Prompt: "I am hiring a Product Manager. Here is the job description: [Paste JD]. Here is a candidate's resume: [Paste Resume]. Compare the resume against the description and provide a summary in three bullet points: 1. Key strengths and direct matches. 2. Potential skill gaps or areas of concern. 3. A list of specific, quantifiable achievements mentioned in the resume."
Generating Targeted Interview Questions
Generic interview questions lead to rehearsed, generic answers. To truly assess a candidate's capabilities, questions must be tailored to the role and the required competencies. You can instruct an AI to generate behavioral or situational questions that probe specific areas like problem-solving, team collaboration, or handling pressure. For technical roles, you can even ask it to create questions related to a specific technology or architectural concept. Example Prompt: "Generate five behavioral interview questions for a 'DevOps Engineer' role. The questions should assess a candidate's experience with infrastructure as code, CI/CD pipeline automation, and handling production incidents. The tone should be for a mid-level professional with 3-5 years of experience."
Personalizing Candidate Outreach
In a market where top engineers receive multiple outreach messages a day, personalization is critical. A generic, copy-pasted message is easily ignored. AI can help craft engaging and personalized messages by referencing specific details from a candidate's profile, such as a recent project on GitHub or a blog post they wrote. This shows genuine interest and dramatically increases response rates. Example Prompt: "Act as a friendly recruiter from a health-tech AI company. Write a 120-word LinkedIn InMail to a passive 'Data Scientist' candidate. Reference their recent Kaggle competition submission on predictive modeling and briefly connect it to the work we are doing in medical diagnostics. End with a low-pressure invitation for a brief chat about their work."
