Shiprocket - AI Engineer - LLM Applications (3-6 yrs)

Description :

Job Title : AI Engineer

Location : Gurugram (On-Site)

Company Overview

Shiprocket is India's leading e-commerce enablement platform, empowering 2,50,000+ sellers to ship
across 24,000+ pin codes in India and 220+ countries. We're building intelligent systems to power the
next generation of logistics and commerce and we're looking for passionate AI Engineers to join
our growing team.

Key Responsibilities :

- Design and develop LLM-based applications including chatbots, summarizers, and intelligent
recommendation systems using frameworks like LangChain and LangGraph.

- Build and maintain RAG pipelines (including Agentic RAG) with vector databases such as Milvus,
Pinecone, or similar.

- Fine-tune and evaluate open-source and proprietary language models (e.g., GPT, LLaMA, Mistral)
on domain-specific datasets.

- Develop and expose AI capabilities via RESTful APIs using Django or FastAPI, with data storage on
MongoDB / SQL databases.

- Implement OCR, document parsing, and text extraction pipelines from PDFs, images, and
structured/unstructured documents.

- Collaborate with product and engineering teams to translate business requirements into scalable
AI solutions.

- Deploy and monitor AI models on AWS (SageMaker, Lambda, EC2, or S3).

- Stay up to date with the latest research in NLP, Generative AI, and ML, and evaluate applicability
to Shiprocket's use cases.

Required Skills & Qualifications :

- 2 to 5 years of hands-on experience in AI/ML or NLP engineering.

- Strong proficiency in Python and libraries such as Pandas, NumPy, and Scikit-learn.

- Hands-on experience with LLM frameworks LangChain, LangGraph, or equivalent.

- Experience building RAG or Agentic RAG systems using vector databases (Milvus, FAISS, Weaviate,
etc.).

- Familiarity with Hugging Face ecosystem Transformers, Datasets, PEFT/LoRA fine-tuning.

- Experience with Django or FastAPI for building production-grade AI APIs.

- Working knowledge of MongoDB and relational databases (MySQL/PostgreSQL).

- Exposure to AWS services for model deployment or data pipelines.

- Good understanding of NLP concepts tokenization, embeddings, semantic search, text
classification.