GenAI Solutions Architecture

Description

Job Summary: We are seeking a solutions architect and developer for modern systems to lead the creation of Generative AI products, working with LLMs, autonomous agents, and RAG. Key Highlights: 1. Lead the creation of Generative AI products 2. Hands-on involvement in architecture and development of modern solutions 3. Focus on LLMs, autonomous agents, RAG, and AI workflows **What We’re Looking For:** =================== We seek a professional specialized in architecture and development of modern solutions, with strong hands-on experience, to lead the creation of Generative Artificial Intelligence products. This profile will be responsible for designing, implementing, and evolving systems based on LLMs, autonomous agents, RAG, and complex AI workflows, as well as defining GenAI technical strategies. Advanced or fluent English is required to collaborate within an international team. **Responsibilities:** ====================== * Define and build system architectures based on LLMs, including RAG pipelines, agents, chatbots, co-pilots, and code assistants; * Collaborate on multi-model orchestration strategies using LiteLLM, including routing, fallback, caching, and rate limiting; * Architect agent systems using LangGraph, covering state graphs, checkpointing, human-in-the-loop, and parallelism; * Implement high-performance RAG pipelines considering embeddings, chunking, rerankers, and hybrid search; * Design memory solutions for agents and chatbots (short-term, long-term, entity memory, conversation summarization); * Define function calling/tool use strategies for integration with APIs, databases, and internal systems; * Develop full-stack applications using .NET (C#), Python, and React/TypeScript, integrating GenAI capabilities; * Build robust APIs to serve AI models with streaming, async processing, and context management; * Implement resilience patterns, including circuit breakers, retry policies, timeout management, and graceful degradation; * Ensure quality through unit tests, integration tests, and LLM-specific evaluations; * Define the company’s GenAI adoption strategy, considering models, providers, and technical trade-offs; * Conduct PoCs and comparative benchmarks across models and approaches; * Establish AI governance policies, monitoring, and guardrails; * Evaluate and implement AI-powered productivity tools; * Provide technical leadership to engineering teams, conducting code reviews, pair programming, and establishing standards; * Mentor developers on GenAI practices and AI tools; * Foster a culture of experimentation through hackathons, tech talks, and rapid prototyping; * Collaborate with Product Managers and stakeholders to translate business needs into technical solutions. **Requirements and Qualifications:** =============================== * Advanced or fluent English. **GenAI & Artificial Intelligence** Proven experience developing LLM-based products in production; * Proficiency in LangChain and LangGraph for chains, agents, and complex workflows; * Experience implementing RAG systems in production, including retrieval and reranking; * Practical experience with LiteLLM or multi-provider abstractions; * Deep expertise in advanced prompt engineering; * Experience with at least two AI providers (Azure OpenAI, AWS Bedrock, Vertex AI/Gemini, Anthropic Claude); * Familiarity with open-source models (Llama, Mistral, Phi) and deployment strategies (vLLM, TGI, Ollama); **Software Engineering** Extensive experience in software development and system architecture; * Solid experience with .NET (C#), Python, and React/TypeScript; * Advanced knowledge of microservices, REST/gRPC APIs, event-driven architecture, and distributed systems; * Experience with cloud computing (Azure, AWS, or GCP) and DevOps/CI-CD practices; * Experience in technical leadership and mentoring engineering teams. **Differentiators:** ================= * Experience designing multi-agent systems (CrewAI, AutoGen, Semantic Kernel); * Experience with model fine-tuning (LoRA, QLoRA, PEFT) and synthetic datasets; * Knowledge of Graph RAG (Neo4j, knowledge graphs) and advanced retrieval techniques; * Experience with LLM observability (LangSmith, Helicone, Phoenix/Arize, Weights & Biases); * Knowledge of Anthropic’s MCP (Model Context Protocol); * Experience with evaluation frameworks (RAGAS, DeepEval, Promptfoo); * Cloud certifications (Azure, AWS, or GCP); **Behavioral Description:** ============================= We are looking for someone who: * Enjoys working in teams and is collaborative in their responsibilities; * Has the courage to challenge themselves and go beyond, embracing new growth opportunities; * Transforms ideas into creative solutions and pursues quality in all daily activities; * Possesses strong problem-solving skills; * Is capable and comfortable working independently and managing their own time; * Is interested in tackling adversarial and innovative technological situations. Big enough to deliver – small enough to care. **#VempraGFT** **#VamosVoarJuntos** **#ProudToBeGFT** With over 35 years in the market and presence in 16 countries, GFT employs more than 12,000 people worldwide, including 4,000 in Brazil. Here, diversity, equality, and inclusion are central, and ensuring a safe workplace for everyone is part of our identity. We promote a culture of development and growth anchored in our Continuous Scaled Learning pillar because we believe intelligent use of technology is the key to success in this digital world. **What We Offer** -------------------- * Multi-benefit card – you choose how and where to use it. * Tuition assistance for undergraduate, graduate, MBA, and language courses. * Certification incentive programs. * Flexible working hours. * Competitive salaries. * Annual performance reviews with structured career plans. * Opportunities for international career progression. * Wellhub and TotalPass. * Private pension plan. * Childcare allowance. * Medical assistance. * Dental assistance. * Life insurance. #VemPraGFT