Architect - Machine Learning - GenAI

Quantiphi; Bengaluru; On-site; INR 4,000,000 - 8,000,000; Full time

Job summary

Quantiphi, Bangalore-based AI solutions company, is hiring an Architect - Machine Learning (AWS) with 12–15 years of hands-on experience to design cloud-native ML architectures.

You will lead end-to-end ML deployments using SageMaker, Bedrock, LangChain and RAG workflows, optimize prompts and models, and collaborate with developers, QA and PMs to deliver robust web applications.

Qualifications

• 12+ years of hands-on experience implementing cloud ML solutions on AWS.
• Hands-on experience with AWS ML services and SageMaker for training, real-time and batch inference.
• Experience building NLP/LLM apps using LangChain and GenAI frameworks.
• Experience with GenAI frameworks such as AWS Bedrock, OpenAI.
• Hands-on experience fine-tuning LLMs (e.g., LLaMA2).
• Experience with Retrieval Augmented Generation (RAG) and vector indexing like OpenSearch/Elasticsearch.
• Strong familiarity with Transformers, BERT, attention models.
• Prompt engineering to optimize few-shot performance for tasks like personalized recommendations.

Skills

AWS SageMaker

LangChain

LLMs & GAIs

GenAI Frameworks

RAG Architecture

OpenSearch/Elasticsearch

Transformers/BERT

Prompt Engineering

AWS Cloud ML Deployment

End-to-End ML on AWS

Tools

SageMaker Studio

Bedrock

OpenSearch

Elasticsearch

LangChain

Job description

Role : Architect - Machine Learning (AWS)

Experience : 12 - 15 Years

Location : Bangalore

Must Have Skills

• 12+ years of relevant hands‑on technical experience implementing, and developing cloud ML solutions on AWS.
• Hands‑on experience on AWS Machine Learning services. Proven experience using AWS SageMaker leveraging different types of data sources, Training jobs, real‑time and batch inference, and processing jobs.
• Good experience developing applications using LLMs with Langchain.
• Must have experience using GenAI frameworks such as AWS Bedrock, OpenAI.
• Must have hands‑on experience fine‑tuning large language models (LLM) and generative AI (GAI), specifically LLaMA2.
• Must have hands‑on experience working with Retrieval Augmented Generation (RAG) architecture and experience using vector indexing such as OpenSearch, Elasticsearch.
• Strong familiarity with higher‑level trends in LLMs and open‑source platforms.
• Should have experience with deep learning concepts: Transformers, BERT, attention models.
• Prompt engineering: Engineer prompts and optimize few‑shot techniques to enhance LLM performance on specific tasks, e.g. personalized recommendations.
• Model evaluation & optimization: Evaluate LLM zero‑shot and few‑shot capabilities, fine‑tune hyperparameters, ensure task generalization, and explore model interpretability for robust web‑app integration.
• Response quality: Collaborate with ML and integration engineers to leverage LLM pre‑trained potential, delivering contextually appropriate responses in a user‑friendly web app.
• Thorough understanding of NLP techniques for text representation and modeling.
• Able to effectively design software architecture as required.
• Experience with at least one workflow orchestration tool: Airflow, StepFunctions, SageMaker Pipelines, Kubeflow, etc. Knowledge of a variety of machine learning techniques (supervised/unsupervised, clustering, decision tree learning, artificial neural networks, etc.) and their real‑world advantages/drawbacks.
• Ability to create end‑to‑end solution architecture for model training, deployment and retraining using native AWS services such as SageMaker, Lambda functions, etc.
• Ability to collaborate with cross‑functional teams such as developers, QA, project managers, and other stakeholders to understand requirements and implement solutions.

Good To Have Skills

• Experience working for customers/workloads in the EdTech domain with relevant use cases.
• Experience with software development.