Junior Retrieval-Augmented Generation (RAG) And LangChain Systems Engineer

What role will you play in the team

As a Junior Retrieval-Augmented Generation (RAG) and LangChain Systems Engineer, you will play a foundational role in building, deploying, and scaling advanced generative artificial intelligence systems. You will collaborate closely with machine learning researchers and data scientists to bridge the gap between static vector databases and dynamic Large Language Models, ensuring our enterprise applications deliver highly accurate, context-aware, and hallucination-free responses for our rapidly expanding client base across various industries.

What you will do

You will design and implement robust data ingestion pipelines, parse unstructured documents, generate dense vector embeddings, and integrate them into scalable vector databases. Additionally, you will fine-tune retrieval mechanisms, orchestrate multi-step LLM workflows using automation tools, and optimize prompt engineering strategies to enhance overall model performance, throughput, and token-efficiency in production environments.

Key responsibility

• Develop and maintain end-to-end RAG pipelines using Python, LangChain, and LlamaIndex for advanced enterprise document querying.
• Integrate OpenAI API and Claude API endpoints securely within microservices architecture using custom APIs and webhooks.
• Configure and manage vector database instances such as Pinecone, Milvus, and PostgreSQL with pgvector extensions.
• Automate AI agent deployment workflows and orchestrate backend tasks utilizing make.com, n8n and Flowise platforms.
• Write rigorous unit tests and benchmark evaluation scripts using scikit-learn and evaluation frameworks to measure retrieval precision and recall.
• Collaborate with cross-functional data teams to ingest unstructured data from diverse sources including PDFs, SQL databases, and cloud storage.
• Monitor API rate limits, optimize prompt engineering templates, and implement caching layers to minimize operational latency and inference costs.

Required Qualification and Skills

• Bachelor's degree in Computer Science, Data Science, Artificial Intelligence, or a related quantitative discipline.
• 0 to 1 years of hands-on professional or academic experience in Python programming and machine learning workflows.
• Demonstrated proficiency with Large Language Models (LLMs), Retrieval-Augmented Generation (RAG), and prompt engineering techniques.
• Familiarity with integration tools like make.com, n8n and Flowise, alongside managing APIs and webhooks.
• Basic understanding of database management systems including SQL, PostgreSQL, and MongoDB.
• Strong analytical mindset with a passion for experimenting with state-of-the-art generative AI frameworks and toolsets.
• Excellent communication skills with the ability to document technical architectures and workflow designs clearly.

Looking to get Placed? Try our Placement Guarantee Plan

Get Hired

Benefits Included

• Competitive entry-level annual compensation package with performance-based quarterly appraisal reviews.
• Comprehensive health insurance coverage including medical, dental, and vision for the employee and dependents.
• Generous paid time off (PTO), public holiday leaves, and flexible bereavement leave policies.
• Continuous learning stipend for online AI/ML courses, certifications, and technical conference registrations.
• Dynamic, collaborative startup work culture with opportunities for rapid career advancement into senior AI roles.
• Well-stocked pantry, complimentary daily catered lunches, and ergonomic workstation setups in the office.

A Day in the Life

Your day typically begins with a morning stand-up to review pipeline metrics and discuss LLM response evaluations with the data science team. You will spend the morning writing Python scripts to parse complex client documents and embedding them into vector databases. After lunch, you will configure an n8n workflow to automate webhook triggers for a new generative AI client feature, followed by debugging an API integration issue with the OpenAI endpoint before wrapping up your tasks.