Role : Agentic AI Engineer

Team : Agentic Automation

Location : Bangalore (Hybrid: Minimum 12 workdays WFO per month)

Experience : 3 - 7 Years (Minimum 3 years relevant experience in AI/ML engineering)

Core Skills : Python (FastAPI), LangGraph, LangChain, Production RAG, AWS (Bedrock), Docker/Kubernetes

Role Overview :

We are looking for a highly skilled Senior AI & Machine Learning Engineer with hands-on experience in Python-based AI/ML engineering and strong expertise across Traditional Machine Learning, Generative AI, and Agentic AI. You will play a critical role in designing, developing, and deploying scalable AI solutions using machine learning models, large language models (LLMs), agentic workflows, cloud technologies, APIs, and modern data platforms.

Key Responsibilities :

- Design and orchestrate LLM-powered applications, production RAG solutions, and stateful agentic workflows using LangGraph and LangChain, including agent state, memory, and tool calling.

- Design, develop, and deploy AI/ML solutions spanning predictive analytics, recommendation systems, forecasting, anomaly detection, and Generative AI use cases.

- Build and operationalize machine learning models, driving the complete lifecycle from data preparation and feature engineering through deployment and monitoring.

- Develop scalable APIs, backend services, and integration components using Python and FastAPI.

- Architect and deploy cloud-native AI solutions on AWS, ensuring security, scalability, reliability, and cost efficiency.

- Build intelligent retrieval and knowledge-grounding capabilities using vector databases, semantic search, and embedding models.

- Implement MLOps practices including CI/CD, model versioning, experiment tracking, monitoring, and automated deployments.

- Collaborate with architects, data engineers, product teams, and business stakeholders to translate business requirements into AI-driven solutions.

- Optimize model performance, application reliability, latency, and operational efficiency in production environments.

- Build evaluation and observability into every agentic release, using test datasets, automated evaluation, tracing, and cost and latency monitoring to prove quality before and after deployment.

- Evaluate emerging AI, machine learning, and agentic framework technologies and drive adoption of relevant innovations.

Must-Have Skills :

- Advanced Python and FastAPI : Production-grade Python (async, typing, testing, packaging) and REST APIs and microservices built with FastAPI.

- LangGraph / LangChain : Hands-on delivery of multi-step, stateful agentic workflows, including graph orchestration, conditional routing, and human-in-the-loop checkpoints.

- Production RAG : RAG systems running in production, covering chunking, embeddings, hybrid search, re-ranking, context engineering, and retrieval-quality tuning.

- Vector databases : Amazon OpenSearch Service, FAISS, Pinecone, Chroma, or equivalent, including index design, metadata filtering, and scaling.

- Agent state and memory : Short-term and long-term memory design, checkpointing, state persistence, and session management for long-running agents.

- Tool calling : Tool and function schemas, structured outputs, error handling, retries, and safe execution of tools with side effects.

- Evaluation frameworks : RAGAS, DeepEval, LangSmith evaluations, or equivalent, with golden datasets, LLM-as-judge scoring, and regression testing in CI.

- Observability : Tracing, logging, and metrics for LLM and agent pipelines using Langfuse, OpenTelemetry, or equivalent, with token, cost, and latency tracking.

- Docker / Kubernetes : Containerizing and deploying AI services on Kubernetes (Amazon EKS preferred), with Git, CI/CD pipelines, and automated testing.

Core Engineering Foundations :

- Minimum 3 years of professional experience in AI/ML engineering, including hands-on experience developing and deploying production-grade AI/ML solutions.

- Demonstrated experience delivering solutions across traditional ML and Generative AI or Agentic AI.

- Experience developing Generative AI and LLM-powered solutions using Amazon Bedrock, and proficiency with AWS services and cloud-native architecture for AI/ML workloads.

- Hands-on experience with machine learning libraries and frameworks such as Pandas, NumPy, scikit-learn, XGBoost, LightGBM, TensorFlow, or PyTorch.

- Sound understanding of supervised and unsupervised learning, feature engineering, model evaluation, hyperparameter tuning, and model explainability.

- Experience working with SQL and NoSQL databases, object storage, and data-processing pipelines.

- Good understanding of scalable software design, API security, and production support.

Good-to-Have Skills :

- MCP and tool registries : Building or consuming Model Context Protocol (MCP) servers and managing governed tool registries for discovery, versioning, and access control.

- AWS Strands Agents or Google ADK : Building agents on AWS Strands Agents SDK or Google Agent Development Kit (ADK).

- Model optimization : Quantization, distillation, parameter-efficient fine-tuning (LoRA/PEFT), prompt caching, and optimized inference serving to cut latency and cost.

- Secure AI engineering : Prompt-injection defence, guardrails (e.g., Amazon Bedrock Guardrails), PII handling, least-privilege tool access, and familiarity with the OWASP Top 10 for LLM Applications.

- Knowledge graphs : Exposure to Graph RAG, entity relationship modeling, and graph databases such as Amazon Neptune or equivalent.

Qualifications :

- B.Tech or Graduation in Computer Science Engineering or relevant field.

What We're Looking For :

- Excellent problem-solving and analytical skills.

- Strong collaborative and communication skills.

- Ability to independently design solutions and work effectively as part of a cross-functional team.

- Ability to explain technical concepts to both technical and business stakeholders.

- Passion for emerging AI technologies and innovation across machine learning and intelligent automation.