ob Description

Role: AI Developer - Agentic AI

Exp: 2-3 Years

Work Mode: 12- 10 pm, Onsite( Mohali, Punjab)
 

Job Role & Responsibilities

• Design, develop, and deploy Agentic AI systems capable of autonomous task execution by integrating reasoning, memory, and tool use to enable intelligent behavior across complex, multi-step workflows.
• Architect intelligent agents that can dynamically interact with APIs, data sources, and third-party tools to accomplish diverse objectives with minimal human intervention.
• Optimize performance of agentic frameworks by enhancing model accuracy, minimizing response latency, and ensuring scalability and reliability in real-world applications.
• Develop reusable, testable, and production-grade code, adhering to best practices in software engineering and modern AI development workflows.
• Collaborate with cross-functional teams, including product managers, designers, and backend engineers, to convert business requirements into modular agent behaviors.
• Integrate Retrieval-Augmented Generation (RAG), advanced NLP techniques, and knowledge graph structures to improve decision-making and contextual awareness of agents.
• Conduct rigorous profiling, debugging, and performance testing of agent workflows to identify bottlenecks and improve runtime efficiency.
• Write and maintain comprehensive unit, integration, and regression tests to validate agent functionality and ensure robust system performance.
• Continuously enhance codebases, refactor existing modules, and adopt new design patterns to accommodate evolving agentic capabilities and improve maintainability.
• Implement secure, fault-tolerant, and privacy-compliant designs to ensure that deployed agentic systems meet enterprise-grade reliability and data protection standards.

Qualification Required:

Bachelor’s degree in computer science, or related field.

Specialization or Certification in AI or ML is a plus.

Technical Expertise:

• 2+ years of hands-on experience in AI/ML/DL projects, with a strong emphasis on Natural Language Processing (NLP), Named Entity Recognition (NER), and Text Analytics.
• Proven ability to design and deploy Agentic AI systems—autonomous, goal-oriented agents that exhibit reasoning, memory retention, tool use, and execution of multi-step tasks.
• Practical expertise in agent architecture, task decomposition, and seamless integration with external APIs, databases, and tools to enhance agent capabilities.
• Skilled in agent prompting strategies, including dynamic prompt chaining and context management, to guide language models through intelligent decision-making workflows.
• Experience with Retrieval-Augmented Generation (RAG) pipelines and generative AI, with a strong focus on optimizing NLP models for low-latency, high-accuracy production use.
• Solid foundation in deep learning methods, recommendation engines, and AI applications within HR or similar domains.
• Exposure to Reinforcement Learning (RL) frameworks and holds relevant certifications or specializations in Artificial Intelligence, showcasing continuous learning and depth in the field.

Minimum skills we look for:

Skills & Expertise (with Agentic AI focus)

• Proven experience in building Agentic AI systems, including autonomous agents capable of multi-step reasoning, memory management, and tool use.
• Expertise in agent design patterns, task decomposition, dynamic planning, and decision-making logic using LLMs.
• Skilled in integrating multi-agent coordination, goal-setting, and feedback loops to create adaptive, evolving agent behavior.
• Strong command over prompt engineering, contextual memory structuring, and tool calling mechanisms within LLM-powered agent workflows.
• Proficiency in managing agent memory (short-term, long-term, episodic) using vector databases and custom memory stores.
• Ability to build autonomous task execution pipelines with minimal human input, combining language models, APIs, and third-party tools.
• Experience with frameworks and orchestration for agent behavior tracing, logging, and failure recovery.

Tools & Technologies – Agentic AI

• Agentic Frameworks: LangChain, CrewAI, AutoGen, AutoGPT, BabyAGI – for building, managing, and orchestrating intelligent agents.
• LLM APIs: OpenAI (GPT-4/3.5), Anthropic (Claude), Cohere, Hugging Face Transformers.
• Memory & Vector Databases: FAISS, Weaviate, Pinecone, Chroma – for embedding-based agent memory and contextual retrieval.
• Prompt Management Tools: PromptLayer, LangSmith – for testing, evaluating, and refining agent prompts and traces.
• RAG & Context Enrichment: LangChain RAG pipelines, Haystack, Milvus.
• Autonomy Infrastructure: Docker, FastAPI, Redis, Celery – for building scalable agent runtimes.
• Observability: OpenTelemetry, Langfuse (or similar) for tracing agent decisions, failures, and success metrics.
• Testing Agentic Behavior: Integration with PyTest + mock APIs/tools to validate autonomous decision logic and fallback strategies.