Company Name: VARITE India Private Limited

About The Client:

Client is one of the world’s leading skilled services firms and the fastest growing Big Four accounting firm

About The Job:

-
- You will architect agent systems, optimize memory management for data and context, and guide cross-functional teams to productionize scalable AI solutions.

Essential Job Functions:
- Architect and build LLM-powered agents with robust tool-use, planning, and multi-turn reasoning.
- Design and implement Agent Memory Management: short/long-term memory, vector stores, episodic/semantic memory, retrieval pipelines, and context optimization.
- Develop and operationalize RAG (Retrieval Augmented Generation) systems, prompt orchestration, and context windows management.
- Own end-to-end AI engineering lifecycle: data pipelines, model integration, evaluation, observability, and MLOps for deployment and scaling.
- Optimize cost, latency, and reliability for production-grade agent workflows.
- Collaborate with Product, Data, and Platform teams; mentor junior engineers.

Qualifications:





Must-Have Skills
- 7–8 years in software/AI engineering with 3+ years hands-on with LLMs and agent frameworks.
- Strong in Python and one or more: TypeScript/Java/Go.
- Experience with LLM frameworks (LangChain, LlamaIndex, DSPy, OpenAI Assistants, Semantic Kernel, AgentQL, etc.).
- Expertise in vector databases (FAISS, Pinecone, Weaviate, Milvus) and embedding models.
- Deep understanding of prompt design, tool calling, function calling, RAG, evaluation (BLEU/ROUGE, grounding, factuality), and guardrails.
- Hands-on with cloud & MLOps (Azure/AWS/GCP), CI/CD, model versioning, observability (prometheus/logging/tracing).
- Experience with data privacy, PII handling, and responsible AI practices.

Good to Have
- Knowledge of fine-tuning/LoRA, distillation, and model quantization.
- Experience building multi-agent systems, planners, and memory graphs.
- Familiarity with streaming, event-driven architec

📌 AI Engineer – LLM (Pune)
🏢 VARITE
📍 Pune