22 days ago
Poland (city not specified)Staff+
Responsibilities
- Lead Generative AI and Agentic AI strategy, roadmaps, use-case identification, and solution proposals.
- Design end-to-end enterprise architectures covering preprocessing, model selection, RAG, agent orchestration, MCP integration, guardrails, training, inference, deployment, and monitoring.
- Define reusable platform patterns for agent orchestration, MCP tools and data integration, shared skills, connectors, memory, state management, human oversight, and observability.
- Evaluate and select models, frameworks, retrieval strategies, and interoperability standards based on data, safety, cost, latency, and computational requirements.
- Collaborate with data scientists and engineers to implement and integrate Generative AI solutions into production environments.
- Optimize model and system performance for speed, accuracy, cost, latency, and resource utilization.
- Assess solution outcomes against success criteria and iterate based on metrics and feedback.
- Collaborate with customer architecture and business teams to define requirements, technical boundaries, and SLAs.
- Mentor junior team members and work with cross-functional technical and business stakeholders.
- Track developments in Generative AI, Agentic AI, MCP, agent skills, models, and related technologies.
Requirements
- Deep expertise in Generative AI, Agentic AI, LLMs, RAG, MCP, multi-agent architecture, and enterprise solution architecture.
- Experience with transformer-based LLMs, diffusion models, multimodal models, GANs, VAEs, and text, code, image, and multimodal generation.
- Experience with prompt engineering, structured outputs, function and tool calling, LangChain, LangGraph, LlamaIndex, and Semantic Kernel.
- Experience designing autonomous and multi-agent systems using ReAct, planning, reflection, tool use, and human-in-the-loop patterns.
- Experience with CrewAI, MAF, OpenAI Agents SDK, Google Agent Development Kit, memory, state management, orchestration, and safe multi-step execution.
- Ability to architect, build, and govern MCP servers and clients and work with MCP tools, resources, and prompts.
- Experience defining modular agent skills, custom tools, connectors, and SKILL.md-style capability modules.
- Expertise in RAG, chunking, embeddings, vector databases, hybrid search, reranking, retrieval evaluation, GraphRAG, and agentic RAG.
- Experience with LLMOps, evaluation, observability, tracing, monitoring, guardrails, red-teaming, and optimization of quality, safety, cost, and latency.
- Understanding of AI governance, security, privacy, bias and fairness, transparency, auditability, and emerging AI regulation.
- Profound understanding of machine learning principles, algorithms, model implementation, performance optimization, and training pipelines.
- Proficiency with Python, TensorFlow or PyTorch, LLM and agent frameworks, cloud AI platforms, vector databases, Docker, Kubernetes, and distributed computing.
- Understanding of data engineering, data pipelines, data management, preprocessing, cleansing, and transformation.
- Excellent communication and collaboration skills, including the ability to explain complex technical concepts to non-technical stakeholders.
- Familiarity with healthcare, finance, entertainment, or other industry domains is beneficial.
Tech Stack
Categories
About Infosys
Infosys is a public IT services and consulting company that builds and runs software systems for large enterprises across industries. It provides consulting, systems integration, cloud and AI services, engineering, and business process outsourcing, and sells platforms such as EdgeVerve's Finacle core banking suite. Founded in 1981 and headquartered in Bangalore, it serves clients in 50+ countries and is listed on the NSE, BSE, and NYSE.