Responsibilities
- Architect and analyze multi-agent AI systems using frameworks like LangGraph, AutoGen, or CrewAI, incorporating reasoning, planning, reflection, tool integration, and memory for autonomous behavior.
- Possess strong knowledge of how multi-modal models function within agent-based architectures.
- Demonstrate expertise in machine learning evaluation metrics such as Precision, Recall, Specificity, and NPV, with experience building evaluators, reward functions, or grading models for reinforcement learning and iterative refinement.
- Apply large language models effectively as evaluators for assessing qualitative and semantic accuracy.
- Build and manage retrieval-augmented generation (RAG) systems with proficiency in embeddings and vector databases including FAISS, Pinecone, Chroma, or Azure AI Search.
- Stay current with advancements in AI model designs, agent frameworks, and innovations in reasoning and information retrieval, showing consistent curiosity and a focus on iterative improvement.
- Maintain a foundational grasp of classical machine learning and core LLM concepts, including transformers, attention mechanisms, and tokenization.
- Understand how context length, embedding spaces, and scaling laws affect model performance, cost, and output quality.
- Fine-tune large language models and multi-modal systems using supervised, reinforcement, or instruction-based methods, including LoRA, PEFT, and TRL.
- Optimize models for efficiency, inference speed, and interpretability without sacrificing accuracy.
- Work closely with platform, product, and application engineers to deliver robust, scalable AI-powered solutions.
- Communicate technical concepts clearly, share insights across teams, and contribute effectively in collaborative environments.
- Apply strong software engineering practices, including modular design and clean code principles, to turn experimental AI concepts into production-grade systems.
Work Arrangement
Remote — Hyderabad
Work Arrangement
Hyderabad
