Software Engineer

Responsibilities:

• Design, build, and deploy AI agents from scratch that can independently reason, plan, make decisions, and execute multi-step tasks.
• Develop agentic workflows involving tool calling, function calling, task decomposition, memory management, and dynamic execution.
• Build robust backend services and APIs to support AI agents using Python, Node.js, or other relevant backend technologies.
• Integrate LLMs such as OpenAI GPT, Anthropic Claude, or other models into agent architectures.
• Implement agent orchestration, state management, context handling, and mechanisms for maintaining conversation and task history.
• Develop multi-agent systems where multiple agents collaborate to solve complex problems, wherever required.
• Build reliable systems with error handling, retries, fallback mechanisms, and safeguards against hallucinations and incorrect tool execution.
• Integrate external APIs, databases, vector databases, and third-party tools to enable agents to retrieve information and perform real-world actions.
• Implement RAG pipelines, semantic search, embeddings, and retrieval mechanisms wherever required.
• Evaluate and optimize agent performance, response quality, latency, reliability, and token costs.
• Deploy and monitor AI agent systems in production, ensuring scalability, observability, and security.
• Take ownership of the complete product lifecycle, from problem definition and architecture to development, deployment, and iteration.

Requirements:

• 3-6 years of professional software engineering experience, with strong expertise in backend development.
• Proven hands-on experience building AI agents or agentic systems from scratch, preferably deployed in production.
• Strong programming skills in Python, Node.js, TypeScript, or similar backend technologies.
• Strong understanding of LLMs, prompt engineering, structured outputs, function calling, and tool use.
• Experience with agent frameworks or orchestration libraries such as LangGraph, LangChain, AutoGen, CrewAI, or equivalent custom-built frameworks.
• Practical knowledge of agent planning, reasoning loops, task execution, state management, and memory.
• Experience integrating external tools, APIs, databases, and services into AI agent workflows.
• Strong understanding of REST APIs, asynchronous processing, microservices, databases, and backend system design.
• Experience with vector databases such as Pinecone, Qdrant, Weaviate, or FAISS.
• Understanding of RAG, embeddings, semantic retrieval, and context-window management.
• Experience deploying, debugging, evaluating, and monitoring AI-powered applications in production.
• Ability to work independently, make architectural decisions, and take ownership in a fast-paced startup environment.

Good To Have:

• Experience building multi-agent architectures and collaborative agent workflows.
• Hands-on experience with Model Context Protocol (MCP), including building MCP servers or clients.
• Experience with AI agent evaluation, observability, tracing, and benchmarking.
• Knowledge of Docker, cloud infrastructure, CI/CD, and scalable distributed systems.
• Experience building AI products from 0 to 1 in a startup or early-stage product environment.