Artificial Intelligence Innovation Specialist

Azul is the largest airline in Brazil in terms of number of destinations, with shares traded on the NYSE. The IT area is not support — it has a strategic seat in the company. Today we operate with tools such as Claude Enterprise, Snowflake, Azure, AWS, GCP, Datadog, ServiceNow, and Microsoft 365 — references of what we use now, not a closed list: what matters is the ability to learn fast and apply the right tool to each problem. We are not looking for an AI theorist, nor a stage futurist. We are looking for someone who wakes up thinking about how to apply, today, what the AI market is already doing — within a real airline, with all its operational and regulatory constraints. Someone who disturbs in a good way: brings benchmarks and concrete cases, builds POCs that work, and does not rest until the experiment becomes a product with measurable ROI. The ideal profile is non-conformist with the status quo and, at the same time, pragmatic enough to separate what is viable from what is premature — and from what is a trend without substance. The difference between this professional and an AI researcher is simple: this one delivers in production.

Responsibilities and assignments

• Continuously monitor the global AI ecosystem — models, frameworks, tools, and use cases in aviation and adjacent sectors (fintech, retail, logistics, health) — filtering what is relevant to Azul's reality.
• Translate market trends into concrete business opportunities, with technical feasibility analysis, implementation effort, and ROI potential for each.
• Build and maintain an active network with suppliers, startups, researchers, and other airlines at the frontier of AI adoption, bringing these connections into Azul.

Experimentation and Innovation

• Design, execute, and document AI POCs with clear hypotheses, objective success criteria, and defined deadlines — without eternal POCs.
• Present benchmarks, cases, and experiment results to leadership and business areas in a way that provokes reflection and accelerates adoption decisions.
• Disturb the organization in a good way: bring external evidence that challenges current practices and creates a sense of urgency for innovation.

Implementation and Value Generation

• Conduct the transition from POC to production, ensuring that successful experiments become scalable solutions with measurable business impact.
• Define and monitor ROI metrics for each AI initiative — revenue generated, cost avoided, hours saved, NPS, among others.
• Work together with business areas (Commercial, Operations, Loyalty, Maintenance, Finance, among others) to identify real problems that AI can solve.
• Reject, with the same conviction, AI applications where AI is not the best solution — prioritizing business results over the technology itself.

Dissemination and Culture

• Be the main AI evangelist at Azul, through workshops, demos, internal communications, and presentations.
• Develop AI maturity in business areas, forming internal partners capable of identifying and co-constructing opportunities.
• Publish or present Azul's cases externally when strategically relevant, positioning the company as a reference in AI in the aviation sector.

Requirements and qualifications

• Practical mastery of LLMs and generative AI in production — not just conceptual: advanced prompt engineering, RAG, fine-tuning, and model evaluation.
• Experience with AI platforms at enterprise scale — today we use Claude Enterprise, Azure, and AWS as reference, but what we seek is transferable know-how, not mastery of a specific tool.
• Knowledge of MLOps and LLMOps: model lifecycle, versioning, monitoring, and evaluation in production.
• Ability to develop and evaluate POCs with code: Python and main AI frameworks (LangChain, LlamaIndex, or equivalents).
• Familiarity with modern data platforms (today we use Snowflake as reference) — does not need to be a data engineer, but needs to know how to consume and orchestrate data.
• Advanced English: reading papers and technical documentation, and relationship with international suppliers and partners.
• Degree in Computer Science, Engineering, Mathematics, Physics, or related areas.
• Postgraduate studies in AI, Machine Learning, or related areas is a differentiator, not a requirement — what counts is the delivery track record.
• Documented AI cases that went from experiment to production with proven ROI.

What We Expect in the First Days

1. Mapping of highest impact opportunities for AI at Azul, with feasibility analysis, estimated effort, and expected ROI — validated with business areas.
2. POCs in execution, with defined success criteria and established evaluation deadlines.
3. POC approved for production, with a scaling plan and defined monitoring metrics.
4. Active AI dissemination program: training or evangelization actions for business areas.
5. Structured and recurring market radar: formal process of monitoring trends with periodic output to leadership.

Work Model and Location

• Hybrid regime: presence around 3 days per week at the office in Alphaville, Barueri.
• Availability for national and international travel (conferences, visits to partners, in loco benchmarks).

An extremely hands-on AI Innovation Specialist, and not just someone with conceptual knowledge. The phrase that best summarizes the position is: “take AI, test fast, prove value, and put into production.” The JD itself makes it explicit that they do not want an “AI theorist” or someone focused only on trends; they want those who transform experiments into products with measurable ROI. JD Especialista AI.docx

The professional I would look for in the market

It is a person who combines technical depth + innovation + business vision + influence. They need to keep up with what is happening most recently in AI, bring benchmarks and external cases, identify what makes sense for the company, and quickly build POCs. JD Especialista AI.docx

The profile is looking for an extremely hands-on AI Innovation Specialist, and not just someone with conceptual knowledge. The phrase that best summarizes the position is: “take AI, test fast, prove value, and put into production.” The JD itself makes it explicit that they do not want an “AI theorist” or someone focused only on trends; they want those who transform experiments into products with measurable ROI. JD Especialista AI.docx

The professional I would look for in the market

It is a person who combines technical depth + innovation + business vision + influence. They need to keep up with what is happening most recently in AI, bring benchmarks and external cases, identify what makes sense for the company, and quickly build POCs. JD Especialista AI.

But there is a crucial point: a POC alone is not enough. The candidate needs to demonstrate cases where they took a solution from experimentation to production, with scale and concrete results — cost reduction, revenue generation, productivity gain, hours saved, NPS, etc. JD Especialista AI.docx

Technically, I would be rigorous on these points

The candidate needs to have practical experience with:

• LLMs and GenAI in production
• RAG, advanced prompt engineering, fine-tuning, and model evaluation
• Python
• LangChain, LlamaIndex, or equivalents
• MLOps and especially LLMOps
• Enterprise AI platforms, such as Azure/AWS/Claude Enterprise or equivalents
• Ability to consume/orchestrate data in modern platforms like Snowflake
• Advanced English. JD Especialista AI.docx

I would place special weight on autonomous agents/Agentic AI, because it appears as a desirable differentiator, especially if they have already put agents into production. JD Especialista AI.docxE behaviorally?

Here is a very important part of this position. It is not enough to be that excellent technician who waits for someone to deliver a perfectly structured project. They want someone provocative, curious, fast, autonomous, and communicative. A person capable of approaching leadership and saying: “I saw this happening abroad, I think we can apply it here, I will test it and come back with numbers.”

At the same time, they need the maturity to say: “This is trendy, but for this problem, it is not worth using AI.”

And they need to talk to both CEOs and executives as well as engineering and operational areas, adapting the language. JD Especialista AI.docx

My reading of the profile in one sentence: they are looking for a “corporate AI builder” with a product and business mindset — someone who researches the market, codes, experiments, builds POCs, influences areas, and takes GenAI to production with proven ROI.

For hunting, I would prioritize candidates with titles such as AI Innovation Specialist, Generative AI Specialist, Lead AI Engineer, GenAI Engineer, Applied AI Engineer, AI Solutions Architect, or AI Innovation Lead. And I would make a rigorous cut: if the resume talks a lot about AI, but does not show GenAI/LLMs in production + POCs + concrete results, I would not consider high alignment. The JD itself reinforces that a delivery track record is worth more than a postgraduate degree and requires documented cases that reached production with ROI.