Ideogram is hiring a Applied Machine Learning in Toronto!
Ideogram is a Toronto-headquartered generative media company building proprietary models and AI-native creative tools. This Applied Machine Learning role sits between research and product: the engineer will turn frontier generative models into dependable features for Ideogram’s first-party app and API. The work spans text-to-image, image-to-text, image enhancement/editing, and multimodal use cases, with an emphasis on rapid experimentation, rigorous evaluation, and production reliability. The engineer will fine-tune and deploy models, design datasets and benchmarks, run failure-mode analyses, and set client success metrics around quality, latency, and cost. Because the role involves frequent client work and 0-to-1 projects, it suits someone who can combine hands-on ML engineering with product judgment and clear communication. Ideogram describes a flat, fast-moving culture with high ownership, mentorship, and collaboration across research, engineering, infrastructure, and product. The position is based in Toronto and the listed perks reference daily in-office meals at the Toronto HQ.
Key responsibilities
- Translate research advances into production ML features for the app and API.
- Fine-tune and deploy generative models across image generation, editing, enhancement, and multimodal workflows.
- Build evaluation corpora, benchmarks, and error analyses to measure model quality and robustness.
- Define client-facing success metrics such as output quality, latency, and inference cost.
- Partner with product, research, infrastructure, and clients to scope and deliver 0-to-1 applied ML projects.
- Contribute to monitoring, safety, reliability, and numerical stability across training and inference.
Key requirements
- At least one year of experience building and shipping machine learning products.
- Strong backend coding skills in Python, with PyTorch or JAX experience.
- Ability to design benchmarks, evaluation metrics, and failure-mode analyses.
- Experience taking ML features from idea through evaluation to production with measurable improvements.
- Comfort curating, labeling, and cleaning datasets and evaluation data.
- Familiarity with modern deep learning concepts such as transformers, diffusion models, and encoders.
- Strong communication and collaboration skills, with comfort working directly with clients and product teams.
- Startup mindset: ownership, urgency, curiosity, and fast learning.
- Seniority: Entry to mid-level (1+ years)
- Work mode: In-office at Toronto HQ
- Key skills: Python, PyTorch, JAX, production ML, model evaluation, benchmark design, error analysis, dataset curation, transformers, diffusion models, backend engineering, client communication
Want more jobs?
Join our mailing list to receive updates about new data science and Applied Machine Learning jobs.
