AI Engineer

Description

**At TRINCA, we don’t hire people—we invite them to build what comes next.** At TRINCA, Artificial Intelligence is not a roadmap trend. It’s about solving real business problems with solutions that work in production, scale under pressure, and deliver measurable value to companies accelerating their digital maturity. We are a strategic tech partner—combining strategy, design, and technology to build what our clients cannot yet envision on their own. And right now, the AI layer is where the most complex and transformative decisions are happening. If you’ve already deployed AI systems into production and understand the difference between an impressive demo and a solution that withstands the real world; if you approach agent architecture as a serious engineering challenge—not just experimentation; if you want to work on projects where impact is concrete and autonomy is real—this role could be your next big step. **Your mission** -------------- Design, develop, and ensure the quality of production-ready Artificial Intelligence solutions, serving as a technical reference in building multi-agent systems, RAG pipelines, and orchestration architectures that translate LLM capabilities into tangible business impact. Here, you won’t build proof-of-concept demos that stay in a drawer. You’ll deliver systems that clients rely on every day. **The challenge** ------------- * Making architectural decisions in a field that changes weekly—balancing what’s solid enough for production with what’s relevant enough for the future * Building multi-agent systems with real reliability: autonomous orchestration, memory, roles, and human\-in\-the\-loop workflows that function even when the scenario deviates from expectations * Ensuring quality in AI solutions through rigorous evaluation frameworks—because "it looks good" is not an acceptable delivery criterion * Translating technical complexity into language non-technical stakeholders can understand and trust—without losing depth in engineering discussions * Navigating distinct business contexts—different projects, different clients, different problems—while maintaining high technical standards across all * Contributing to the evolution of AI practice at TRINCA, serving as an internal reference in a domain still being collectively built **What you’ll do daily** ------------------------------------- ### **AI Architecture and Development** * Design and implement agent and multi-agent system architectures, defining roles, tools, memory strategies, and autonomous orchestration flows * Develop in Python with expertise in LLM orchestration frameworks such as LangChain, LangGraph, LlamaIndex, CrewAI, or similar—choosing the right tool for each problem, not the most popular one * Build and evolve RAG pipelines with deliberate decisions around chunking, embeddings, vector databases, and semantic or hybrid search strategies * Integrate LLMs with external APIs, databases, and tools via function calling—ensuring secure, efficient, and traceable orchestration standards * Apply and refine prompt engineering and context engineering practices with technical rigor—not as recipes, but as informed decisions grounded in context and model behavior ### **Quality and Evaluation** * Implement AI solution evaluation frameworks—RAGAS, DeepEval, LangSmith, Braintrust, or similar—as a mandatory part of the delivery cycle, not an optional step * Write and maintain unit tests and code reviews with the perspective of someone who knows AI systems fail in ways traditional tests don’t cover * Document architectural decisions and implementation standards clearly, ensuring knowledge isn’t siloed within a single person ### **Interoperability and Ecosystem** * Explore and apply emerging agent interoperability protocols—such as Model Context Protocol (MCP) and Agent\-to\-Agent (A2A)—in real projects * Contribute DevOps and CI/CD insights to ensure AI solutions are deployed, monitored, and evolved with the same discipline applied to any software system ### **Collaboration and Impact** * Actively participate in agile rituals—refinements, planning, retrospectives—with technical contributions that help the team make better decisions * Collaborate with developers, designers, PMs, and QAs to ensure AI solutions integrate into the product with technical integrity and coherent user experience * Adopt a consultative stance with clients—understanding the business problem before proposing any technical solution **What we’re looking for in you** -------------------------- ### **Essential requirements** * 5+ years of software engineering experience, including at least 2 years dedicated to developing and delivering AI solutions in production—not research or isolated experimentation * Proficiency in Python with hands-on experience using LLM orchestration frameworks: LangChain, LangGraph, LlamaIndex, CrewAI, or equivalents * Solid experience designing and developing agent and multi-agent system architectures, with practical exposure to autonomous orchestration and human\-in\-the\-loop workflows * Experience building end-to-end RAG pipelines—from chunking to retrieval—with understanding of trade-offs among different strategies * Familiarity with AI evaluation frameworks and the practice of measuring solution quality and performance before considering delivery complete * Knowledge of integrating LLMs with external APIs and databases via function calling * Experience with agile methodologies and version control using GitHub * Knowledge of DevOps and CI/CD applied to AI projects ### **Behavioral competencies** * Clear communication with both technical and non-technical audiences—knowing when to simplify without sacrificing precision * A genuinely consultative and proactive attitude—engaging with the problem before being asked and proposing paths forward, not just options * Authentic collaboration in multidisciplinary teams, openness to feedback, and willingness to co-create * Commitment to continuous learning in a field that doesn’t wait for those who stop studying ### **Nice-to-have differentiators** * Familiarity with emerging protocols such as Model Context Protocol (MCP) and Agent\-to\-Agent (A2A) * Experience with Microsoft Azure and cloud ecosystem in AI projects * Experience in consulting firms or in projects involving multiple clients and distinct business contexts * Experience developing solutions for the insurance sector or other regulated industries **TRINCA’s approach to artificial intelligence** --------------------------------------------- * **Embrace the challenges:** AI in production differs from AI in demos. Here, we expect those who confront the real-world complexity of systems—hallucinations, latency, orchestration failures, difficult evaluation—with method, responsibility, and a determination to solve it once and for all * **Connect authentically:** the best AI solution doesn’t emerge from an isolated engineer. It emerges from someone who understands the client’s business, collaborates with product and design, and builds *with* the team—not *for* the team * **Deliver with pride:** here, "works in a notebook" is not delivery. We deliver evaluated, documented, monitored, and production-ready systems. If you hold AI to the same rigor you apply to any other critical software system, you’re in the right place **Why this could be your next great opportunity** --------------------------------------------------------- * Work on AI applied to real business problems in high-complexity projects for large enterprises—not disconnected experiments lacking impact * Shape TRINCA’s AI practice at its formative stage—meaning your technical decisions define the path forward, rather than merely following an established pattern * Direct access to leadership and clients involved in decision-making—no layers of approval diluting the impact of your technical work * Real autonomy to choose tools, propose architectures, and influence how the company builds AI solutions * A growth and international expansion context—with global market ambitions and projects broadening the horizon for everyone on the team * A culture of continuous learning, with chapters, authentic technical exchanges, and a team that treats AI seriously as engineering *We Build What's Next.* **And the next future can begin with you.**