About the Role

We are looking for enthusiastic and curious engineering students to design and build next-generation Workflow-Driven AI Agents. Unlike standard chat applications that rely on one-shot prompts, this role focuses on creating agile, autonomous agentic systems that use Large Language Models (LLMs) to reason, plan, execute actions with external tools, maintain state, and continuously refine output.

Key Responsibilities

- System Architecture & Workflow Design: Build structured AI workflows utilizing agentic design patterns such as ReAct, Plan-and-Execute, and multi-tool orchestration.
- Autonomous Task Execution: Implement agents capable of breaking down user goals into logical steps, executing tool calls (APIs, web search, code execution, database lookups), observing results, and handling edge cases gracefully.
- LLM Integration: Integrate commercial or open-source LLMs (e.g., OpenAI, Anthropic Claude, LLaMA, Mistral) into workflow engines.
- Context & State Management: Maintain persistent context and execution states across multi-step, multi-turn reasoning loops.
- Code & Documentation Quality: Deliver clean, modular, and well-documented source code (GitHub), along with architecture diagrams and demo videos.

Required Technical Skills

- Programming Languages: Proficiency in Python (strongly recommended) or JavaScript/TypeScript/C++.
- LLM & AI Frameworks: Hands-on experience or familiarity with AI Agent frameworks such as LangChain, LangGraph, CrewAI, or AutoGen (or building custom agent loops from scratch).
- API & Tool Integration: Experience with REST APIs,



web scraping, JSON processing, and calling external software/tools from Python.
- Prompt Engineering & Reasoning Patterns: Understanding of structured prompting, tool/function calling with LLMs, and agentic loops (Plan $rightarrow$ Act $rightarrow$ Observe $rightarrow$ Respond).
- Version Control: Git and GitHub best practices.

Preferred / Good-to-Have Skills

- Familiarity with vector databases (Chroma, Pinecone, FAISS) and Retrieval-Augmented Generation (RAG).
- Experience deploying local open-source models (e.g., via Ollama, Hugging Face, vLLM).
- Basic understanding of state machines and workflow orchestration.

Candidate Qualifications

- Currently pursuing a Bachelor’s or Master’s degree in Engineering (CSE, ISE, ECE, AI/DS, or equivalent quantitative discipline).
- Strong analytical, problem-solving, and algorithmic skills.
- Ability to demonstrate past projects, hackathon prototypes, or open-source contributions.

Deliverables Expected During the Project / Role

1. Functional AI Agent: A fully working implementation performing a dedicated task (e.g., Autonomous Research Agent, Task Planner, Code Helper, Data Analysis Assistant, or Customer Support Agent).
2. Codebase & Architecture: A clean public/private GitHub repository with a detailed README.md including a system architecture diagram and setup instructions.
3. Product Demonstration: A 2–5 minute video demonstration showing the agent’s reasoning steps and execution flow.

How to Apply: Send your resume or portfolio directly to [email protected]

📌 AI Agent Development Intern / Developer (India)
🏢 Techvruk
📍 India