ML Engineer, Fine Tuning

Slack is looking for a Staff Machine Learning Engineer with deep expertise in model training and fine tuning to join our ML team. You'll design, train, and ship NLP models that power core product experiences — from summarization and search ranking to generative AI features used by millions daily. This role is hands‑on: you'll work at a low level with training frameworks, optimize model architectures, build fine tuning pipelines, and own the full lifecycle from experiment to production.

• Design and execute fine tuning strategies for large language models and other deep learning architectures tailored to Slack's NLP tasks (summarization, ranking, classification, generation).
• Own the model training lifecycle end‑to‑end: data curation, training infrastructure, hyperparameter optimization, evaluation, deployment and monitoring.
• Build and maintain scalable fine tuning training pipelines on GPU infrastructure.
• Brainstorm with Product Managers, Designers and Frontend Engineers to conceptualize and build new features for our large (and growing!) user base.
• Produce high‑quality results by leading or contributing heavily to large multi‑functional projects that have a significant impact on the business.
• Mentor other engineers and deeply review code.
• Improve engineering standards, tooling, and processes.

• 5+ years of hands‑on experience training and fine‑tuning deep learning models in NLP (or a closely related domain like speech, IR, or multimodal).
• 5+ years of experience with common deep learning frameworks like PyTorch, Tensor Flow, JAX, etc.
• Track record of shipping fine‑tuned models to production that serve real users at scale — not just research prototypes.
• Experience with functional or imperative programming languages: PHP, Python, Ruby, Go, C, Scala or Java.
• An analytical and data driven mindset, and know how to measure success with complicated ML/AI products.
• Led technical architecture discussions and helped drive technical decisions within the team.
• The ability to write understandable, testable code with an eye towards maintainability.
• Strong communication skills and you are capable of explaining complex technical concepts to designers, support, and other specialists.

• Expertise with recommendation systems or search.
• Familiarity with model optimization for inference (quantization, pruning, speculative decoding, compilation via Torch Script/Tensor
  
  RT/ONNX).
• Experience with retrieval‑augmented generation and hybrid retrieval/generation systems.
• Broad experience across NLP, ML, and Generative AI capabilities.
• Knowledge of using multiple data types in RAG solutions including structured, unstructured, and knowledge graphs.

Salesforce is an equal opportunity employer and maintains a policy of non‑discrimination with all employees and applicants for employment. What does that mean exactly? It means that at Salesforce, we believe in equality for all. And we believe we can lead the path to equality in part by creating a workplace that’s inclusive, and free from discrimination. Know your rights: workplace discrimination is illegal.

Any employee or potential employee will be assessed on the basis of merit, competence and qualifications – without regard to race, religion, color, national origin, sex, sexual orientation, gender expression or identity, transgender status, age, disability, veteran or marital status, political viewpoint, or other classifications protected by law. This policy applies to current and prospective employees, no matter where they are in their Salesforce employment journey.

It also applies to recruiting, hiring, job assignment, compensation, promotion, benefits, training, assessment of job performance, discipline, termination, and everything in between. Recruiting, hiring, and promotion decisions at Salesforce are fair and based on merit. The same goes for compensation, benefits, promotions, transfers, reduction in workforce, recall, training, and education.

In the United States, compensation offered will be determined by factors such as location, job level, job‑related knowledge, skills, and experience. Certain roles may be eligible for…