Generative AI Engineer
First seen by Alion on Sep 28, 2026.
About the job :
About CoffeeBeans Consulting :
CoffeeBeans is a tech-driven software consulting company that helps businesses solve complex problems using modern data, AI, and engineering solutions. We blend deep technical expertise with a product mindset to build scalable, intelligent, and high-impact solutions across industries. Our data science team works on end-to-end delivery from exploration and modeling to GenAI application development and deployment.
Experience : 3 - 5 Years
Location : Hyderabad | Work mode : Work from Office (5 Days)
Role Overview :
As an L2 Data Scientist, you will play a hands-on role in delivering production-grade ML and GenAI-powered applications. You are expected to independently take ownership of data science components within client projects, contribute to solutioning and design, and mentor junior team members. You will work across a range of use cases such as personalization, fraud detection, intelligent automation, RAG pipelines, and LLM-based assistants. This role is ideal for someone who has proven experience in both traditional ML and an emerging understanding of LLMs and generative AI applications.
Key Responsibilities :
ML & Data Science :
- Own and deliver ML model development, tuning, and evaluation for client-facing projects.
- Lead exploratory data analysis, data preprocessing, and feature engineering with minimal supervision.
- Build models using appropriate ML techniques (classification, regression, clustering, recommendation) and ensure performance meets business expectations.
- Contribute to experimentation frameworks and model reproducibility best practices.
GenAI & LLM Applications :
- Design and prototype GenAI solutions using LLMs (e.g., OpenAI, Claude, Mistral, Llama).
- Hands-on experience with AgenticAI.
- Build RAG pipelines, prompt templates, few-shot learning prompts, and evaluation mechanisms for GenAI systems.
- Integrate LLMs with APIs, vector databases (e.g., Pinecone, FAISS, Weaviate), and context providers.
- Contribute to benchmarking, safety, and cost-performance trade-offs in LLM app development.
Product & Engineering Collaboration :
- Collaborate with engineering teams to take models from experimentation to deployment (batch/real-time).
- Assist in building APIs and data pipelines needed for productionizing models.
- Contribute to technical documentation, explainability reports, and client presentations.
Team & Growth :
- Mentor junior data scientists and review code/model design.
- Stay current with advances in ML and GenAI to inform solution design and share knowledge internally.
- Participate in discovery and solutioning phases with clients alongside tech leads and PMs.
Required Skills & Qualifications :
- Bachelors or Masters degree in Computer Science, Data Science, Engineering, Statistics, or a related field.
- 2 - 5 years of hands-on experience in applied data science, including both ML model development and GenAI-based solutioning.
- Strong command over Python and ML libraries (scikit-learn, XGBoost, LightGBM, etc.).
- Experience working with LLM APIs (OpenAI, Cohere, Claude, etc.) and frameworks (LangChain, LlamaIndex, or similar).
- Hands-on experience with prompt engineering, RAG workflows, and evaluating LLM outputs.
- Proficiency in SQL and data wrangling tools (pandas, NumPy).
- Experience working with REST APIs, Git, and cloud environments (AWS/GCP).
Good-to-Have Skills :
- Experience with deploying models via FastAPI, Docker, or serverless platforms.
- Knowledge of MLOps tools (MLflow, DVC) and monitoring frameworks.
- Experience with embeddings, vector databases, and similarity search.
What You Can Expect at CoffeeBeans :
- Work on real-world AI/ML problems across verticals.
- Be part of a fast-moving team delivering end-to-end ML & GenAI apps.
- Collaborate with experienced engineers and PMs in a flat and open culture.
- Opportunities to lead, mentor, and influence tech direction.
Skills
Generative AI, Artificial Intelligence, AI Integration, LLM, RAG, Data Science, SQL, Data Pipeline, VectorDB