Description :

A leading consulting firm operating in the Enterprise Generative AI and Large Language Model (LLM) services sector, delivering production-grade LLM solutions, retrieval-augmented systems, and custom generative AI products for enterprise clients across domains. The team focuses on building secure, scalable, low-latency inference services and automating model lifecycle workflows for on-prem and cloud deployments.

Position : LLM Engineer - On-site (India). We are hiring an experienced LLM engineer to design, fine-tune, and deploy LLM-based solutions that power search, summarization, agents, and domain-specific assistants.

Role & Responsibilities :
- Design, fine-tune, and validate LLMs for production use-cases : Instruction tuning, supervised fine-tuning, and parameter-productive tuning (LoRA/adapters)
- Implement retrieval-augmented generation (RAG) pipelines : Embeddings, vector search, chunking, and context assembly for high-recall responses.
- Optimize inference for latency and cost : Quantization, model pruning, batching, and deployment with optimized runtimes (CUDA,



Triton, bitsandbytes where applicable).
- Build backend services and APIs to serve LLM inference and orchestration using containerized deployments (Docker/Kubernetes) and CI/CD pipelines.
- Collaborate with product, data engineering, and ML teams to integrate LLMs into production flows, monitor model performance, and set up automated retraining/rollbacks.
- Create reproducible training pipelines, implement evaluation suites, and produce documentation and runbooks for model governance and observability.

Skills & Qualifications :

Must-Have :
- 4+ years of hands-on experience working with LLMs or advanced NLP models in production contexts.
- Proficiency in Python for ML engineering and model development.
- Experience with PyTorch and Hugging Face Transformers for training and fine-tuning.
- Practical experience implementing RAG and vector search using tools like FAISS or similar

vector

📌 LLM Engineer (Hyderabad)
🏢 Zorba consulting
📍 Hyderabad