Engineering Manager AI in Bogotá at DEUNA

DEUNA is looking for an Engineering Manager to lead our AI/ML engineering team — the group building the intelligent systems behind payment routing optimization, authorization rate improvement, and our AI-powered digital workforce products. You will manage a team of engineers (currently ~7-8) spanning ML, backend, and platform work, while staying close enough to the technical details to guide architecture decisions, unblock your team, and represent the group's roadmap to product and leadership.

This is a hands-on management role: you won't be writing production code day-to-day, but you need enough depth in ML systems, backend services, and AI/LLM workflows to make good calls on architecture, review technical approaches, and coach engineers through hard problems. You'll own delivery for the team — scoping, costing, and sequencing initiatives — while building a low-tech-debt, high-quality engineering culture.

• 8+ years in software engineering, including 2-3+ years in a people management role leading engineers.

• You have significant experience building and shipping backend and/or ML systems at scale, ideally with exposure to fintech, payments, or another regulated, latency-sensitive domain.

• You've managed engineers before, hiring, developing, and retaining a team, and you know how to balance people development with delivery pressure.

• You have hands-on familiarity with modern AI/ML systems: model training and serving, LLM-powered workflows (agents, RAG, orchestration), or similar — enough to have a real technical conversation with your team and challenge their thinking when needed.

• Practical exposure to LLM-based systems in production (agents, RAG, or AI workflow orchestration).

• Payments, fintech, or another regulated-industry background is a strong plus, but not required.

• You communicate clearly and proactively, both with your team and with cross-functional partners in product and operations.

• You're comfortable in a fast-moving startup environment — priorities shift, and you can re-scope and re-communicate without losing the team's trust.

• You're driven by self-improvement and push the people around you to grow as well.

Team & People Leadership

• Manage and grow a team of AI/ML and platform engineers, including hiring, performance development, and career growth.

• Coach engineers through technical design decisions, code and architecture reviews, and hard trade-offs.

• Set the engineering bar for the team: code standards, testing strategy, and CI/CD practices.

• Drive accurate costing and delivery estimates for initiatives, and keep the team accountable to commitments.

Technical Ownership

• Guide architecture for ML model lifecycle work (training, evaluation, monitoring, retraining) and LLM-powered workflows (agent orchestration, RAG pipelines, vector DB integrations).

• Oversee inference services supporting live payment routing, ensuring they meet strict latency and reliability requirements.

• Ensure the team's AWS infrastructure, CI/CD, and observability practices (dashboards, tracing, on-call runbooks) meet a high bar.

• Apply sound judgment on PCI-DSS and data-handling requirements across anything touching payment data.

Cross-functional Collaboration

• Translate product vision into an executable technical roadmap with clear timelines and trade-offs.

• Partner directly with product, operations, and modeling leadership to keep feedback loops short and priorities aligned.

• Represent the AI/ML engineering team's progress and blockers to leadership.

• Backend / Platform: Go, Python, gRPC & REST APIs, event streaming, distributed systems

• Cloud & Infra: AWS (ECS/EKS, Terraform, RDS/Aurora, S3), hybrid/on-prem deployment

• AI / ML Stack: PyTorch/TensorFlow, XGBoost/scikit-learn, MLflow/W&B, model monitoring

• LLMs & Agents: LangGraph/LangChain, RAG, vector DBs, prompt engineering, LLM evaluation

• Payments Domain: PCI-DSS awareness, tokenization patterns, PSP integrations, routing/auth rate optimization

• Observability: Prometheus/Grafana, OpenTelemetry, structured logging, on-call runbooks