AI Research Internship

The post describes a startup working on AI-powered agents and automation. The intern's job is to help the team understand the agent landscape, test what the tools can do, and feed that into prototypes.

• Research generative AI agents, LLM-based agents and autonomous agent frameworks in depth.
• Go through recent papers, blogs, GitHub projects and industry news on AI agents.
• Benchmark agent frameworks and toolkits against each other on performance, capabilities and real-world use.
• Help the team prototype simple agent workflows and simulations.
• Write up findings as reports, documentation and presentations that the team uses to share knowledge internally.
• Look for gaps and possible use cases for AI agents in particular industries, with e-commerce, finance and customer support given as examples.
• Assist with proof-of-concept code in Python, JS or another suitable language, to try out different agent architectures.

The post also says interns work closely with the founders and tech leads.

The conditions are fairly strict, so read them before you spend time on an application.

• You should be a Computer Science Engineering student, or pursuing a degree in CS, AI/ML, Data Science or a related field.
• You must be from Gurgaon and available for a full-time, in-office internship.
• You should be able to start between 16th Oct'26 and 20th Nov'26.
• The eligibility list says 6 months. Another part of the post mentions a 3 to 6 month internship.
• Women restarting their careers can also apply.

Skills asked for: Python, Machine Learning, AI and generative AI development. Also wanted: a real interest in LLMs and multi-agent systems, the ability to read and summarise research papers, and good analytical, writing and presentation skills. Knowledge of OpenAI APIs, LangChain or LlamaIndex is a bonus, as are prompt engineering or model fine-tuning.

• Pick two or three recent papers on LLM agents and write a one-page summary of each. Reading and summarising papers is part of the job, so this is direct practice.
• Build a small agent in Python with any framework you can access, and note what worked and what broke. A short write-up of that comparison shows the kind of benchmarking the role needs.
• If you have tried OpenAI APIs, LangChain or LlamaIndex, say so plainly and link to what you made. Hands-on work counts as a bonus here.
• Practise explaining a technical idea in a few clear slides. The role involves regular reports and presentations.
• Think of one industry, such as customer support, and sketch where an AI agent could help and where it would not. That shows the use-case thinking the role involves.