Marina Wyss, Senior Applied Scientist at Twitch, discusses how AI is reshaping machine learning careers for 2026. She highlights a shift from coding to planning and evaluation, the rising importance of AI engineering as a stepping stone, and strategies for networking and interview preparation. She emphasizes continuous learning, communication skills, and building real-world projects to stand out in a competitive job market.
Summarized by Podsumo
Machine learning roles are shifting from 60% coding to more planning and scoping, with AI code assistants accelerating delivery.
AI engineering is a high-demand, accessible entry point; machine learning engineering is now a superset requiring both traditional and GenAI skills.
A master's degree is helpful but not mandatory; networking and real-world portfolio projects (e.g., volunteer work) can bypass resume filters.
Evaluation skills are critical for non-deterministic AI systems; systematic prompt tracking and golden sets ensure quality.
Career success requires comfort with ambiguity, proactive business impact thinking, and speaking up in meetings to influence decisions.
"If there's a thousand applicants, they might just cut based on advanced degrees and not even see you otherwise. — Marina Wyss"
"It's completely a fool's errand if you're trying to keep up with everything. It's not possible. — Marina Wyss"
"For me at least I can't just walk into an interview and nail it but what I can do is spend 100 hours thinking of every question they're ever going to ask. — Marina Wyss"