Full Stack Developer

TEKsystems•Bellevue, WA

2d•Onsite

About The Position

We’re looking for a Full Stack Developer OR Machine Learning Engineer to build end-to-end systems around ML workflows, with a focus on enabling LLM/model training and fine-tuning. This is a full-stack ML engineering role: you’ll work across data (SQL + performance fundamentals), backend services in Python, ML training workflows, and TypeScript UIs that help researchers and engineers run experiments, evaluate results, and ship improvements. This role is ideal for a builder who has owned systems end-to-end—often seen in startup environments—though strong enterprise experience is also welcome.

Requirements

8+ years of software engineering experience (or equivalent), with ownership of complex systems in production.
Demonstrated experience building end-to-end products spanning data + backend + frontend.
Strong coding experience in Python and TypeScript (production-level in both).
Strong exposure to ML workflows (beyond calling model APIs): data preparation, training runs, evaluation, iteration loops, and deployment/rollout constraints.
Hands-on experience fine-tuning or training ML models (LLMs preferred; other deep learning experience acceptable).
Solid understanding of SQL performance fundamentals (query design, indexing concepts, debugging slow queries, etc.).
Ability to operate in ambiguity, lead technical decisions, and drive delivery with minimal oversight.

Nice To Haves

Experience building internal ML developer tooling: experiment tracking, eval harnesses, model registry, dataset versioning/lineage, guardrails.
Familiarity with training infrastructure concepts (schedulers/orchestrators, checkpointing, failure recovery, profiling).
Experience with LLM quality evaluation approaches (offline eval suites, regression tracking, human/LLM-as-judge patterns).
Startup-style generalist background (breadth, speed, pragmatic decision-making).

Responsibilities

Build and evolve ML workflow systems: dataset creation/curation, job orchestration, experiment tracking, evaluation, and model/version management.
Contribute hands-on to model training activities, such as running fine-tunes, improving training stability, debugging training/data issues, and iterating on quality.
Develop backend services and APIs (Python-first) that integrate data, training, and evaluation components into reliable tooling.
Build TypeScript-based UIs (and supporting services) that enable users to launch runs, compare experiments, inspect metrics, and debug pipeline failures.
Work deeply with data access patterns and performance: write efficient queries and reason about SQL schema/performance concepts (including indexing fundamentals and tradeoffs).
Improve engineering quality for ML systems: reproducibility, testing, CI/CD for ML workflows, monitoring/alerting, and safe rollouts.
Partner closely with ML researchers, product partners, and infra to translate needs into clear deliverables and milestones.