AI/ML Engineer

Bechtel; Reston (VA); Hybride; USD 123 000 - 179 000; Temps partiel

Résumé du poste

Bechtel is seeking an AI/ML Engineer to serve as the AI quality authority within the AI Program in Reston, VA. The role focuses on establishing engineering standards, implementing validation pipelines, and ensuring governance for AI/ML deployments across Bechtel infrastructure projects.

The position is part-time telework with at least three days of in-person work per week at the Reston office. You will collaborate with data science, software, and project teams to deliver trustworthy AI solutions

Qualifications

• Bachelor’s or Master’s degree in Data Science, CS, ML, Statistics, or a related quantitative engineering field.
• 5+ years of progressive technical experience in Data Science, ML Engineering, or AI Quality/MLOps roles.

Responsabilités

• Establish, maintain AI engineering standards, design testing requirements and development best practices across the AI Program.
• Design, build, and maintain automated testing frameworks and evaluation pipelines (MLflow, Databricks) for ML models and LLM apps.
• Validate AI models and agentic systems against defined criteria for performance, accuracy, safety, and reliability.
• Architect and oversee HITL validation mechanisms, including reviewer roles, criteria, and escalation paths.

Connaissances

Python

ML frameworks

MLOps

Model evaluation

Documentation

Prompt engineering

LangChain

NeMo Guardrails

Formation

Bachelor’s or Master’s in Data Science/CS/ML/Statistics

Outils

MLflow

Databricks

TruLens

Ragas

LangChain

NeMo Guardrails

Description du poste

Requisition ID: 297512

• Relocation Authorized: None
• Telework Type:Part-Time Telework
• Work Location: Reston, VA
• Salary Range: 123,400 - 178,500 annually (salary range is based on market data; final salary offered is determined by education, experience, and qualifications of the applicant.)

Extraordinary teams building inspiring projects:

Since 1898, we have helped customers complete more than 25,000 projects in 160 countries on all seven continents that have created jobs, grown economies, improved the resiliency of the world's infrastructure, increased access to energy, resources, and vital services, and made the world a safer, cleaner place.

Differentiated by the quality of our people and our relentless drive to deliver the most successful outcomes, we align our capabilities to our customers' objectives to create a lasting positive impact. We serve the Infrastructure; Nuclear, Security & Environmental; Energy; Mining & Metals, and the Manufacturing and Technology markets. Our services span from initial planning and investment, through start-up and operations.

Core to Bechtel is our Vision, Values and Commitments. They are what we believe, what customers can expect, and how we deliver. Learn more about our extraordinary teams building inspiring projects in our Impact Report.

Job Summary:

The AI Program is a strategic cross-business initiative delivering AI capabilities at scale across Bechtel Infrastructure as part of the wider Execution Transformation Initiative. The AI Program relies on rigorous technical assurance to ensure that AI models, agents, and coding solutions deployed across Bechtel meet quality, safety, and accuracy standards. As AI capabilities scale from pilots to production, structured validation frameworks, coding standards, and Human-in-the-Loop (HITL) review processes are essential to maintaining trust in AI-driven outputs.

The AI/ML Engineer will serve as the technical quality authority within the AI program. This role is responsible for establishing AI engineering, coding, and testing standards; designing and implementing validation and evaluation pipelines for machine learning models, generative AI applications, RAG solutions, and agentic workflows; and implementing technical guardrails that safeguard solution integrity.

"This position is designated as part-time telework per our global telework policy and will require at least three days of in-person attendance per week at the assigned office or project. Weekly in-person schedules will be determined by the individual and their supervisor, in consultation with functional or project leadership" #LI-RM1

Major Responsibilities:

• Technical Quality Authority: Establish, maintain, and assess compliance with AI engineering standards, software design patterns, testing requirements, documentation expectations, and development best practices across the Infrastructure AI Program.
• Automated Evaluation Pipelines: Design, build, and maintain automated testing frameworks and evaluation pipelines using technologies such as MLflow and Ragas for traditional machine learning models, LLM-based applications, RAG solutions, and single-agent or multi-agent workflows.
• Model & Agent Assurance: Validate AI models and agentic systems against defined, measurable, and risk-based criteria for performance, accuracy, reliability, groundedness, safety, security, and responsible AI before production deployment or broader rollout.
• HITL Architecture: Architect and oversee risk-based Human-in-the-Loop validation mechanisms, including reviewer roles, review criteria, approval thresholds, sampling strategies, exception handling, evidence requirements, and escalation paths. Provide documented technical assurance recommendations for high-risk AI workflows to support formal governance and deployment decisions.
• Risk & Drift Mitigation: Identify model degradation, data drift, concept drift, bias, AI-specific security risks, and other quality issues. Define required corrective actions, collaborate with accountable solution owners on remediation, and verify that identified issues have been appropriately addressed.
• Platform & Governance Alignment: Collaborate with the Data Solutions Architect and platform teams to integrate automated tests and technical quality gates into approved MLOps platforms, Databricks workflows, Azure DevOps processes, and CI/CD pipelines. Ensure alignment with applicable architecture, data governance, cybersecurity, and responsible AI requirements.
• Benchmarking & Reporting: Maintain comprehensive technical documentation, model validation reports, and benchmark tracking to support program governance and auditability.
• Technical Guidance: Provide coaching, technical code reviews, and training to engineering and data delivery teams on AI testing methodologies, guardrails, and quality assurance.

Education and Experience Requirements:

• Education: Bachelor’s or Master’s degree in Data Science, Computer Science, Machine Learning, Statistics, or a related quantitative engineering field.
• Experience: 5+ years of progressive technical experience in Data Science, Machine Learning Engineering, or AI Quality/MLOps roles, with demonstrated hands‑on experience deploying or evaluating production AI/ML systems.

Required Knowledge and Skills:

• Core AI/ML Engineering & Coding: Advanced proficiency in Python and hands‑on experience developing maintainable AI or machine learning solutions using relevant frameworks and libraries such as scikit-learn, PyTorch, TensorFlow, Pandas, or NumPy.
• AI Evaluation & Quality Frameworks: Proven experience designing quantitative evaluation suites, test cases, and quality gates for LLMs, RAG architectures, and agentic solutions (preferably hands‑on with tools like Databricks LLM Evaluation, MLflow, TruLens, or Ragas).
• Generative AI Concepts: Deep understanding of Transformer models, vector embeddings, prompt engineering, agent orchestration, and retrieval‑augmented generation (RAG).
• MLOps & Pipeline Integration: Direct experience integrating automated model tests, evaluation pipelines, and assurance gates into CI/CD workflows and MLOps platforms such as Databricks or Azure Machine Learning, using Git‑based version control and development practices.
• AI/ML Metrics Rigor: Strong background evaluating both statistical ML metrics (precision/recall, ROC‑AUC, MAE) and generative AI metrics (faithfulness, answer relevance, semantic similarity, toxicity).
• Documentation & Communication: Ability to translate complex statistical and model validation metrics into technical governance reports and present findings clearly to leadership.

Preferred Qualifications:

• Experience with agentic frameworks (e.g., AutoGen, LangChain, CrewAI, or Microsoft Copilot Studio).
• Familiarity with responsible AI frameworks, guardrail libraries (e.g., NeMo Guardrails), and bias detection methodologies.
• Knowledge of EPC industry workflows, engineering data formats, or heavy asset project delivery environments.

Total Rewards/Benefits:

For decades, Bechtel has worked to inspire the next generation of employees and beyond! Because our teams face some of the world's toughest challenges, we offer robust benefits to ensure our people thrive. Whether it is advancing careers, delivering programs to enhance our culture, or providing time to recharge, Bechtel has the benefits to build a legacy of sustainable growth. Learn more at Bechtel Total Rewards

Diverse teams build the extraordinary:

As a global company, Bechtel has long been home to a vibrant multitude of nationalities, cultures, ethnicities, and life experiences. This diversity has made us a more trusted partner, more effective problem solvers and innovators, and a more attractive destination for leading talent.

We are committed to being a company where every colleague feels that they belong-where colleagues feel part of "One Team," respected and rewarded for what they bring, supported in pursuing their goals, invested in our values and purpose, and treated equitably. Click here to learn more about the people who power our legacy.

At Bechtel, our employees enjoy a competitive total rewards package that includes comprehensive medical, dental, and vision plans, along with optional disability and supplemental insurance options, generous paid time off (160 hours annually, accrued 6.16 hours per pay period), nine paid holidays, paid parental leave, discretionary bonuses, and a well‑designed 401K plan with matching and profit‑sharing components

Bechtel is an Equal Opportunity Employer. All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, sexual orientation, gender identity and expression, age,national origin, disability, citizenship status (except as authorized by law),protected veteran status, genetic information, and any other characteristic protected by federal, state or local law. Applicants with a disability, who require a reasonable accommodation for any part of the application or hiring process, may e-mail their request to [e-mail]