ZoftSolutions - Machine Learning Engineer - LLM/RAG

Key Responsibilities :

- Design, develop, and deploy ML models for Agentic AI use cases.

- Work with AWS AI/ML ecosystem (SageMaker, Bedrock, Lambda, Step Functions, S3, DynamoDB, Kinesis).

- Preprocess and engineer features from structured, unstructured, and streaming data.

- Collaborate with data engineers to ensure high-quality, well-curated training datasets.

- Implement LLM fine-tuning, embeddings, and retrieval-augmented generation (RAG) pipelines.

- Evaluate and optimize models for accuracy, performance, scalability, and cost-efficiency.

- Integrate models into production applications and APIs.

- Work with MLOps teams to automate training, testing, deployment, and monitoring workflows.

- Perform experimentation, A/B testing, and model validation to ensure reliability.

- Document experiments, pipelines, and best practices for reproducibility.

Required Skills & Qualifications :

- 3 to 6 years of experience in ML engineering (adjust based on seniority).

- Strong programming skills in Python (NumPy, Pandas, Scikit-learn, PyTorch, TensorFlow).

- Solid understanding of ML lifecycle (data preprocessing, training, evaluation, deployment).

- Experience with AWS services for ML (SageMaker, Lambda, ECS/EKS, Step Functions, Bedrock).

- Familiarity with large language models (LLMs), NLP, and embeddings.

- Strong knowledge of APIs and microservice deployment.

- Experience with ML pipeline orchestration (Airflow, Kubeflow, MLflow, or similar).

- Understanding of data versioning, experiment tracking, and model registry.

- Proficiency in SQL/NoSQL databases and vector databases (Weaviate, Pinecone, FAISS).