Forward Deployed Engineer - LLM Post-training
Reflection4 months ago
San Francisco, CA, USA or New York, NY, USAMid Level
H1B Sponsor
Responsibilities
- Fine-tune open-weight models for customer-specific use cases by preparing datasets, configuring training runs, and iterating based on evaluations
- Build and maintain evaluation infrastructure, including evaluation suites, test sets, baselines, and improvement measurements
- Clean, format, and assess raw customer data while identifying adversarial or noisy samples
- Build reproducible data pipelines for training data
- Debug training and inference issues by interpreting loss curves and training dynamics
- Support deployment of fine-tuned models across public cloud, VPC, and on-premises environments
- Help ensure inference performance and reliability in production
- Develop fine-tuning playbooks, evaluation benchmarks, and best practices
- Collaborate with customers and research teams to translate requirements and advance model capabilities
Requirements
- Hands-on experience fine-tuning language models, including preparing datasets, running training loops, evaluating results, and shipping a fine-tuned model
- Familiarity with supervised fine-tuning, preference optimization, reinforcement fine-tuning, DPO, RLHF, or similar techniques
- Understanding of evaluation methodology, training graphs, benchmark overfitting, and model improvement measurement
- Comfort working with GPUs, compute management, and debugging training failures
- Strong Python software engineering fundamentals and experience writing clean, reproducible code
- Experience with data pipelines and version control for datasets and experiments
- At least 3 years of engineering experience with meaningful exposure to applied ML or ML engineering
- Ability and interest to work directly with customers, understand user needs, and translate domain requirements into training strategies
- Self-starter with strong ownership who can work effectively in a fast-paced startup environment
Benefits
- Comprehensive medical, dental, vision, life, and disability insurance
- Fully paid parental leave for all new parents, including adoptive and surrogate journeys
- Financial support for family planning
- Paid time off
- Relocation support
- Daily lunch and dinner provided
- Regular off-sites and team celebrations
- Salary and equity compensation are offered, but amounts are not specified
Tech Stack
Categories
Forward DeployedML Engineering
About Reflection
Reflection is a research lab making intelligence open and accessible for everyone to use, customize, and build on. Our team previously built frontier LLMs at labs like DeepMind, OpenAI, and Anthropic. We believe AI should be built in the open, with transparent research and collaborative development. That means giving enterprises, governments, and sovereign entities true ownership and control of AI that performs at the highest level. Our mission: make intelligence open and accessible to all.