Senior Machine Learning Engineer

We’re looking for a Senior Machine Learning Engineer to lead LLM-powered application development-from prototype to production-on AWS. You’ll design robust ML/LLM services that power search, recommendations, copilots, and workflow automation in Brightly’s platform, partnering closely with product, data, and engineering teams. Responsibilities and skill expectations reflect current industry practice for senior ML/LLM engineers, including end-to-end model lifecycle ownership, production-grade code, and MLOps.

What you’ll do

• Build LLM applications: Design and implement RAG pipelines, prompt orchestration, tools/agents, safety/guardrails, and evaluation harnesses; instrument for latency, cost, and quality. (Guided by current LLM engineer role practices.)

• Own the ML lifecycle: Data curation, feature engineering, training/fine-tuning (LoRA/QLoRA), A/B testing, deployment, monitoring, and continuous improvement of models and prompts.

• Productionize on AWS: Ship scalable services on EKS/ECS/Lambda; leverage SageMaker, Bedrock, EMR, MSK, Step Functions; apply observability (CloudWatch/OpenTelemetry) and cost controls. (Duties aligned to modern AWS ML roles.)

• MLOps & governance: Establish CI/CD for models (MLflow/Kedro/SageMaker Pipelines), model/version registries, data and prompt lineage, evaluation gates, and responsible-AI controls. (Aligned with contemporary MLOps templates.)

• Partner across Brightly: Translate asset-management use cases into ML/LLM solutions; collaborate with product managers and UX to ship customer-visible features that measurably improve reliability, safety, and sustainability.

• Perform Exploratory Data Analysis (EDA) on structured, semi-structured, and unstructured datasets to identify patterns, correlations, feature importance, and data quality issues. (Consistent with ML engineer responsibilities to analyze data before model development.)

• Conduct deep research on asset-related, operational, and domain-specific datasets to understand root causes, trends, and predictive signals.

• 8-10 years total software/ML engineering experience, with 2+ years building and operating ML systems in production.

• 1+ years hands-on LLM application development (e.g., RAG, fine-tuning, prompt engineering, evaluators/guardrails, agentic workflows) using packages such as Langchain and Langgraph.

• AWS proficiency (3+ years): Strong with core services (EKS/ECS, Lambda, S3, DynamoDB/RDS, Step Functions, IAM) and ML stack (SageMaker, Bedrock or HF on AWS). (Representative AWS ML role skills.)

• Modeling & frameworks: Python, PyTorch, Hugging Face ecosystem; vector stores (e.g., OpenSearch, PGVector, Pinecone), embeddings, retrieval, and evaluation metrics for NLP/LLMs. (In line with senior LLM roles.)

• MLOps: CI/CD for ML, model registries, experiment tracking, telemetry/monitoring, automated retraining; Docker/Kubernetes, GitHub Actions/GitLab CI. (Current MLOps expectations.)

• Data engineering fluency: ETL/ELT, streaming/batch (Spark/Flink), data quality and governance controls for ML.

Nice to have

• Experience with distributed training (FSDP, DeepSpeed), RLHF, or Inferentia/Trainium optimization.

• Exposure to sustainability/asset/intelligent operations domains.

• Familiarity with security & compliance for ML systems in enterprise environments. (Frequently included in senior ML roles.)

How you’ll work

• Pragmatic and product-oriented: You bias to measurable outcomes and iterate quickly with stakeholders. (Modern senior ML role framing.)

• Engineering excellence: You write production-quality Python, design reliable APIs/services, and uphold testing/observability standards. (Common duties in senior templates.), * Bachelor’s in CS/EE/Math or related field (Master’s preferred) or equivalent practical experience. (Typical for senior ML roles.)