Becoming a Research Engineer at a Big LLM Lab 18 Months of Strategic Career Dev (www.maxmynter.com)

A former ML engineer describes an 18-month, deliberate campaign that landed them a research engineer role at Mistral, a well-funded frontier model lab. After deciding in April 2024 to step up, they alternated strategic and tactical work: brushing up CS fundamentals (distributed systems, data structures, algorithms), heavy LeetCode prep, then resigning to focus full-time on applications in Jan 2025. A pivotal strategic phase was a three-month stint at Recurse Center (remote cohort), where they learned Rust, made 15 PRs to mature OSS projects (ruff and uv), and co-authored a research paper with the AI Safety Institute — all of which strengthened their portfolio. By mid-2025 they ran a disciplined application playbook (60 touchpoints, 40 companies), used referrals and batched interview processes, and secured a verbal offer in August before signing in September. The account matters to the AI/ML community because it’s a practical blueprint for breaking into highly competitive LLM labs: combine compounding, high-effort strategic moves (deep portfolio projects, tenure, research) with tactical readiness (mock interviews, coding drills, targeted outreach). Technical takeaways include the value of systems-level skills (Rust, distributed systems), contributing to high-quality open-source code, and publishing research to signal rigor. Equally important are goal clarity, network cultivation, and process design (batching applications, low-stakes practice) — a repeatable strategy for engineers aiming to move into top-tier ML research/engineering roles.