Tezo is a new generation Digital & AI solutions provider, with a history of creating remarkable outcomes for our customers. We bring exceptional experiences using cutting-edge analytics, data proficiency, technology, and digital excellence.
Experience: 4 to 7+ years (with a minimum of 2+ years of hands-on experience in Generative AI/LLMs)
Location: Hyderabad
Responsibilities:
- Design, build, and deploy production-grade Generative AI applications and architectures utilizing state-of-the-art Large Language Models (LLMs).
- Develop complex multi-step prompt engineering templates, chain-of-thought instructions, and system guardrails for application workflows.
- Collaborate closely with backend, cloud, and product teams to integrate AI models seamlessly into scalable microservices.
- Architect, optimize, and maintain Retrieval-Augmented Generation (RAG) pipelines to cleanly merge enterprise data with language models.
- Establish automated RAG evaluation and LLM output validation frameworks to mitigate hallucinations, ensure factual correctness, and monitor semantic drift.
- Mentor junior team members and champion best practices in ML/LLMOps.
Must have:
- Programming & Core ML: Strong, production-level proficiency in Python and classical Machine Learning frameworks (Scikit-Learn, PyTorch, or TensorFlow).
- Gen AI Architecture: Proven experience building enterprise-grade RAG pipelines, managing data chunking strategies, embedding generation, and vector databases (e.g., PGVector, Pinecone, Milvus, or FAISS).
- Prompt Engineering: Deep expertise in advanced prompt engineering strategies, few-shot learning, and structured outputs (e.g., JSON validation via Pydantic).
- LLM Evaluation & Validation: Hands-on experience setting up evaluation workflows using frameworks like G-Eval, Ragas, or TruLens to score model outputs on relevance, correctness, and safety.
Nice to Have:
- Experience with orchestration frameworks like LangChain, LlamaIndex, or LangGraph (Agentic AI).
- Familiarity with fine-tuning techniques (LoRA, QLoRA) and deploying small/open-source models (Llama, Mistral).
- Experience with cloud platforms (Azure Preferred) and containerization tools (Docker, Kubernetes).
- Knowledge of hybrid search strategies and re-ranking algorithms (e.g., Cohere Rerank).
Qualification:
- Bachelor’s or Master’s degree in Computer Science, Data Science, Artificial Intelligence, or a closely related technical field.
- Excellent communication, problem-solving, and team leadership skills