About the role:
Ciklum is looking for a Senior Machine Learning Engineer to join our team in Brazil or Argentina. We are a custom product engineering company that supports both multinational organizations and scaling startups to solve their most complex business challenges. With a global team of over 4,000 highly skilled developers, consultants, analysts and product owners, we engineer technology that redefines industries and shapes the way people live.
As a Senior Machine Learning Engineer, you'll become a part of a cross-functional development team engineering experience of tomorrow. The company builds software for commercial real estate. This isn't general cloud engineering, you will actually built agentic-platform infrastructure before, not just traditional cloud architecture.
Model Design & Delivery
- Design, train and ship production ML models for real estate and facilities decisions, such as forecasting for space, occupancy, headcount, rent and finances.
- Own each model end to end: problem framing, data and features, training, validation, deployment, monitoring, retraining, and cost and latency budgets.
- Serve models as versioned, documented APIs that product features, agents and other systems call.
- Keep the serving layer portable across cloud platforms, for example by using containers.
- Contribute to embedding and retrieval work, and measure retrieval quality on real customer documents.
- Contribute to synthetic data pipelines, so teams can develop and test without production customer data.
- Measure how closely the synthetic data matches the real data, how well models trained on it perform, and whether it leaks private data.
Evaluation, Quality & Trust
- Own the evaluation of every model you ship: backtests, held-out datasets, metrics that fit the problem, and regression suites in CI that gate releases.
- Build them on the shared evaluation framework that the platform team provides.
- Find the root cause of quality failures, whether data drift, leakage, a feature defect, flawed ground truth, a retrieval miss or a model regression. Fix the layer that is at fault.
- Monitor production models for accuracy, drift, silent degradation and cost.
- Retrain or roll back based on evidence.
- Own model-level safety behavior: keep customer data within each tenant’s access rules, handle PII correctly, test for privacy leakage, and disclose uncertainty clearly.
Cross-Functional Collaboration
- Partner with Product to turn accuracy thresholds and confidence disclosure into shipped behavior.
- Agree on API contracts, accuracy bars and failure behavior with the teams that build AI agents on your models.
- Work with Platform Engineering on deployment and with Data Platform on the curated datasets and features your models read.
- Feed model performance and usage data into the warehouse, so model quality is measurable next to product analytics.
- Support the domain teams that build on your models, contribute to the shared quality standard for models and agents, and mentor engineers on ML practice.
Requirements
- You have taken at least one ML model from problem framing to production, and kept it accurate through retraining and drift.
- Typically 5+ years of professional engineering experience and a degree in Computer Science, Statistics, Mathematics, Engineering or a related field, or equivalent experience.
- Python and SQL, including the ML stack (scikit-learn, PyTorch or equivalent, pandas or equivalent) and the service stack (FastAPI, Pydantic or equivalent).
- Testing, code review, CI/CD and production ownership.
- You use AI coding assistants and verify their output.
- Hands-on experience with ML evaluation: time-aware validation and backtesting against simple baselines, holdout design, metric selection, interval coverage and calibration, error analysis, and using results to gate releases.
- This is central to the role.
- Production experience with forecasting models, with a working knowledge of statistics and probability.
- You can quantify uncertainty, for example with prediction intervals.
- Model serving and MLOps: reproducible deployment, monitoring and retraining, for example with experiment tracking, a model registry, containers, and batch or online inference.
- Judgment about failure modes such as data leakage, drift, distribution shift, bias, privacy leakage and silent degradation.
- You can tell acceptable failures from unacceptable ones, and explain model behavior and uncertainty to Product and to people who are not engineers.
Desirable
- Embeddings and retrieval systems, including how to measure retrieval quality.
- Synthetic data generation, privacy attack evaluation (for example membership inference), and familiarity with differential privacy.
- Geospatial or location data modeling.
- Experience building models that LLM agents call as tools, with an agent framework such as LangGraph and LLM tracing tools.
- Batch pipeline orchestration (Airflow, Step Functions or equivalent) and self-hosted model inference.
- Experience in real estate, facilities, finance or another data-heavy enterprise domain.
What’s in it for you?
- Care: your mental and physical health is our priority. We ensure comprehensive company-paid medical insurance and mental health programs, 5 undocumented sick-leave days per year.
- Tailored education path: boost your skills and knowledge with our regular internal events (meetups, conferences, workshops), Udemy license, language courses and company-paid certifications.
- Growth environment: share your experience and level up your expertise with a community of skilled professionals, locally and globally.
- Long-term employment with 20 working-days paid vacation and local bank holidays.
- Flexibility: 100% remote work mode.
- Opportunities: we value our specialists and always find the best options for them. Our Internal Mobility Program helps change a project if needed to help you grow, excel professionally and fulfill your potential.
- Global impact: work on large-scale projects that redefine industries with international and fast-growing clients.
- Welcoming environment: feel empowered with a friendly team, open-door policy, informal atmosphere within the company and regular team-building events.
About us:
At Ciklum, we are always exploring innovations, empowering each other to achieve more, and engineering solutions that matter. With us, you’ll work with cutting-edge technologies, contribute to impactful projects, and be part of a One Team culture that values collaboration and progress. As we expand into Latin America, every Ciklumer is helping to shape our story. Collaborate with seasoned experts and make a global impact backed by two decades of industry leadership. Explore, empower, engineer with Ciklum!