About the Role
We are seeking a talented and driven Generative AI (GenAI) Engineer with 2–6 years of experience to join our growing AI innovation team in Ahmedabad or Pune. In this role, you'll work on cutting-edge projects involving Large Language Models (LLMs) and Small Language Models (SLMs), with a focus on Retrieval-Augmented Generation (RAG), contextual similarity, and transformer-based architectures like BERT.
Key Responsibilities
- Develop and optimize RAG pipelines, including modules for sorting, filtering, ranking, and re-ranking of retrieved data.
- Work with both closed-source and open-source LLMs/SLMs to solve real-world business problems.
- Implement Chain of Thought (CoT) reasoning to enhance model interpretability and output accuracy.
- Design and apply contextual similarity, graph-based RAG, and corrective RAG techniques to improve model relevance and trustworthiness.
- Fine-tune and utilize BERT and other transformer models for embeddings, classification, and NLP tasks.
- Develop scalable APIs and microservices using Flask to integrate GenAI functionality into applications.
- Research and prototype emerging GenAI capabilities for integration into existing systems.
- Collaborate cross-functionally with product managers, ML engineers, and software developers to deliver robust GenAI solutions.
Required Skills & Qualifications
- 2–6 years of hands-on experience in Python and AI/ML development.
- Deep understanding of LLMs and SLMs, including implementation and integration strategies.
- Experience with Retrieval-Augmented Generation (RAG) architecture and retrieval optimization techniques.
- Proficiency in contextual similarity, vector-based search, and semantic relevance models.
- Strong understanding of Chain of Thought (CoT) prompting and reasoning techniques.
- Experience working with Graph RAG and Corrective RAG for enhanced model accuracy.
- Familiarity with BERT and other transformer-based models for NLP use cases.
- Practical experience building APIs using Flask or similar frameworks.
Preferred Qualifications
- Experience with vector databases (e.g., FAISS, Pinecone, Weaviate).
- Familiarity with LangChain, Haystack, or LLM orchestration tools.
- Experience deploying GenAI solutions on cloud platforms (AWS, Azure, GCP).
- Knowledge of prompt engineering, tokenization, and conversational memory.