Senior AI Engineer

Circuitry.ai; Hyderabad; Hybrid; INR 4,000,000 - 7,000,000; Full time

Job summary

Circuitry.ai is seeking a Senior AI Engineer to design, build, and deploy agentic AI systems and Generative AI solutions in production. You will craft AI agents and multi-agent workflows with CrewAI, LangGraph, and related frameworks, integrating RAG, knowledge graphs, and ML models to power AI Advisors across enterprise workflows.

The ideal candidate brings hands-on experience with GenAI, embeddings, vector databases, and production-grade APIs, with emphasis on performance, cost, and

Qualifications

• 4+ years of software engineering with strong Python, including 2+ years building production GenAI, RAG or agent applications.
• Hands-on experience building agentic workflows with CrewAI, LangGraph, AutoGen or the OpenAI Agents SDK is a plus.
• Strong understanding of LLM APIs, prompt and context engineering, tool calling, MCP and structured outputs.
• RAG experience: embeddings, vector databases, hybrid search and reranking.

Responsibilities

• Design and build AI agents and multi-agent workflows using CrewAI, LangGraph and similar frameworks.
• Implement tool calling, MCP, agent-to-agent communication, planning, memory and human-in-the-loop approvals.
• Orchestrate multi-step workflows that route tasks across agents, tools and enterprise systems.
• Develop scalable APIs and microservices, and support them in production.
• Monitor and optimize latency, token usage and inference cost.

Skills

Python

GenAI / LLMs

RAG / embeddings

CrewAI / LangGraph

FastAPI

Docker

AWS

Knowledge graphs

Tools

Neo4j

OpenAI API

Langfuse

Hugging Face

Kubernetes

Job description

Position: Senior AI Engineer

Experience: 4+ Years

Location: Hyderabad

Work Model: Work from office & Hyrbid Model

Employment Type: Full-Time

About the Role

We are seeking a Senior AI Engineer to design, build and deploy agentic AI systems and Generative AI solutions in production. You will build AI agents and multi-agent workflows with frameworks such as CrewAI. You will combine them with RAG, knowledge graphs and machine learning to power AI Advisors and automate enterprise workflows. The ideal candidate keeps pace with the fast-moving AI landscape and brings new tools and techniques into production.

Key Responsibilities

Agentic AI

• Design and build AI agents and multi-agent workflows using CrewAI, LangGraph and similar frameworks.
• Implement tool calling, MCP, agent-to-agent (A2A) communication, planning, memory and human-in-the-loop approvals.
• Orchestrate multi-step workflows that route tasks across agents, tools and enterprise systems.

Generative AI and RAG

• Build Generative AI applications using OpenAI, Azure OpenAI, Claude, Gemini and similar foundation models.
• Develop RAG pipelines: embeddings, chunking, hybrid search, reranking and context engineering.
• Design and maintain vector database and knowledge graph (e.g. Neo4j) solutions.

AI/ML

• Build and apply ML models for classification, clustering, similarity, ranking and prediction.
• Evaluate model, answer and agent quality using test sets, metrics and LLM-as-judge, then find root causes of failures and drive fixes.
• Evaluate and adopt new models, frameworks and tools as the field evolves.

Engineering and Production

• Develop scalable APIs and microservices, and support them in production.
• Monitor and optimize latency, token usage and inference cost.
• Implement guardrails, permission-aware data access and responsible AI practices.
• Build and maintain CI/CD pipelines for AI deployment and monitoring.
• Collaborate with Product, Data Engineering and Business stakeholders on use cases and architecture.

Required Skills & Qualifications

• 4+ years of software engineering with strong Python, including 2+ years building production GenAI, RAG or agent applications.
• Hands‑on experience building agentic workflows with CrewAI. LangGraph, AutoGen or the OpenAI Agents SDK is a plus.
• Strong understanding of LLM APIs, prompt and context engineering, tool calling, MCP and structured outputs.
• RAG experience: embeddings, vector databases, hybrid search and reranking.
• Solid AI/ML fundamentals: supervised and unsupervised learning, NLP, transformers and evaluation metrics, using scikit‑learn, PyTorch or Hugging Face.
• Experience with knowledge graphs or graph databases (Neo4j/Cypher preferred).
• Backend development with FastAPI, Docker and AWS (Lambda, EKS/ECS, S3).
• Eval design and regression testing for LLM and agent systems.
• Keeps current with the latest AI models, frameworks and research, and can judge what is ready for production.

Nice to Have

• Agent SDKs and platforms: OpenAI Agents SDK, Claude Agent SDK, Google ADK, AWS Bedrock AgentCore.
• A2A protocol, agent memory stores, and LLM observability (Langfuse, OpenTelemetry, Datadog).
• LLM fine-tuning (LoRA, QLoRA, PEFT).
• Kubernetes, Terraform/IaC, GitHub Actions, Azure.
• OAuth 2.1, RBAC, secrets management.
• A GitHub portfolio of agent projects, open-source contributions or AI publications.

Thanks

Satya Puruma