We are seeking an AI Engineer with strong Python skills and hands-on experience building production-grade ML/LLM-enabled systems. You will design advanced RAG pipelines, develop agentic workflows using LangGraph/LangChain, and deploy scalable AI services via FastAPI. This role also includes computer vision and multimodal capabilities—image preprocessing with OpenCV/PIL and working with image generation models such as Stable Diffusion and FLUX for enterprise use cases.

Key Responsibilities

• Design and implement RAG pipelines end-to-end: ingestion, chunking, embedding, retrieval, reranking, grounding, and evaluation for knowledge-intensive applications.

• Design and implement end-to-end image generation pipelines using Hugging Face models and the Diffusers ecosystem, covering dataset/input preparation, prompt/workflow orchestration, inference serving, safety/quality checks, and production monitoring.

• Build autonomous/agentic AI workflows using LangGraph and frameworks like LangChain (tool calling, routing, state/memory, multi-step orchestration, guardrails).

• Develop and maintain Python REST APIs for AI services using FastAPI, including request/response schemas, validation, error handling, API versioning, and documentation.

• Design, train, and validate supervised and unsupervised ML models for business use-cases (classification, regression, ranking).

• Productionize models: containerize ML artifacts/services with Docker, define CI/CD pipelines, and deploy services to cloud-based infrastructure.

• Build and maintain reliable data ingestion and feature engineering pipelines using SQL and Python data-processing libraries.

• Implement model monitoring and quality controls: evaluation metrics, offline/online testing, A/B testing, drift detection, latency/cost monitoring, and rollback strategies.

• Integrate AI/ML services into backend systems; collaborate with engineers and product owners to deliver end-to-end solutions.

• Work on computer vision pipelines: preprocessing, augmentation, and transformation using OpenCV and PIL; integrate CV modules into ML workflows.

• Document experiments, maintain reproducible workflows, and contribute to engineering best practices and knowledge sharing.

Requirements

Must-Have Skills & Qualifications

• 5+ years of professional experience building and deploying ML/AI solutions.

• Strong programming skills in Python with solid software engineering practices.

• Hands-on experience with ML/deep learning frameworks: PyTorch, TensorFlow, scikit-learn.

• Strong applied statistics and ML fundamentals (evaluation, regularization, feature engineering, error analysis).

• Strong SQL skills and experience building data/feature pipelines.

• Experience with LLMs and NLP techniques for enterprise applications (prompting, embeddings, RAG patterns, evaluation).

• Strong experience building REST APIs using FastAPI.

• Experience with containerization using Docker.

• Excellent communication skills. Ability to clearly explain technical concepts to non-technical stakeholders, write clean documentation, and collaborate effectively across engineering, data, and product teams.

Preferred Skills

• Experience with AWS (or Azure/GCP) and deploying inference services in cloud environments.

• Kubernetes for orchestration, autoscaling, and production deployments.

• MLOps practices: CI/CD for ML, experiment tracking, model registry, automated evaluation, monitoring, and governance.

• Experience with agentic frameworks: LangGraph, LangChain, AutoGen (or similar), including tool integration and multi-agent orchestration.

• Experience with RAG infrastructure components (vector databases, embedding models, rerankers) and evaluation strategies.

• Inference optimization (batching, caching, concurrency tuning) and cost-performance trade-offs.

• Familiarity with enterprise security/compliance (PII handling, access controls, auditability).

Nice-to-Have

• Practical knowledge of OpenCV and PIL for image processing integrated into ML pipelines.

• Experience implementing image generation workflows using Stable Diffusion and FLUX (pipelines, prompt engineering, safety filtering, evaluation).

• Experience building multimodal applications (text + image) and deploying them as reliable services.

• GenAI testing and evaluation experience: building/using evaluation harnesses for RAG/agentic systems, defining quality metrics, regression testing prompts/workflows, and monitoring quality drift in production.