Senior AI Full Stack Engineer - Hybrid (GC/ US Citizens Eligible to Obtain Clearance)

Swingtech•Washington, DC

13d•Hybrid

About The Position

The Senior AI Full Stack Engineer will serve as the program’s lead hands-on technical executor for the development, integration, testing, deployment, enhancement, and operational support of secure AI-enabled applications. The individual will implement the standards and architecture patterns established by the Lead AI Engineer or Government architects and will have direct responsibility for core application, RAG pipeline, testing, security, configuration, deployment, and operational deliverables.

Requirements

At least five years of hands-on experience in one or more of the following: AI/ML engineering, AI-enabled application development, Data engineering, Cloud engineering, Software engineering.
Experience must include direct development, integration, testing, or operational support of AI-enabled systems.
At least one year of hands-on experience with one or more of the following: Generative AI, LLM integration, Retrieval-Augmented Generation, Model evaluation, Agentic AI, Production AI systems.
Strong proficiency in Python.
Experience developing AI-enabled applications, APIs, services, integrations, and automation.
Experience designing and deploying RAG pipelines, semantic retrieval, embeddings, vector databases, cloud-based AI services, and AI workflow orchestration.
Knowledge of AI testing, model evaluation, Responsible AI, AI security controls, and secure enterprise or Federal cloud environments.
Experience conducting AI-specific adversarial testing and documenting remediation.
Experience operating in FedRAMP-authorized environments and applying the FISMA Moderate NIST SP 800-53 control baseline.
Experience developing AI-enabled interfaces conforming to Section 508 and WCAG 2.1 Level A and AA.
Experience supporting release management, technical documentation, system monitoring, troubleshooting, and production operations.
Must hold at least one relevant AI, Generative AI, machine learning, data engineering, or cloud AI certification.
Must be a U.S. citizen or legal permanent resident.
Must be eligible to obtain and maintain a favorable Tier 1 background investigation, formerly NACI.

Nice To Haves

Duties involving privileged or higher-risk access may require a Moderate- or High-Risk background investigation.
Must satisfy required identity verification, confidentiality, security, training, and onboarding requirements before receiving access.

Responsibilities

Design, develop, integrate, deploy, operate, and enhance AI-enabled applications and services.
Build and maintain LLM integrations, RAG and advanced RAG pipelines, embeddings, semantic retrieval, vector search, agentic workflows, and AI-enabled automation.
Develop secure AI applications, APIs, backend services, user interfaces, dashboards, integrations, and workflow components.
Provide hands-on development, integration, testing, troubleshooting, deployment, and operational support.
Implement model evaluation, hallucination reduction, retrieval-quality tuning, grounding validation, monitoring, and performance optimization.
Maintain knowledge-base ingestion and refresh processes, embedding regeneration, vector-database configurations, and retrieval-relevance monitoring.
Implement and enforce environment controls across DEV, TEST, STAGING, and PROD.
Ensure changes complete applicable unit testing, automated security scanning, model evaluation, accessibility testing, and TEVV gates before production.
Manage secrets using approved FedRAMP-authorized secret-management capabilities and prevent hardcoded or shared credentials.
Support MLOps workflows, model and prompt versioning, drift detection, retraining support, CI/CD, release documentation, and production monitoring.
Conduct code reviews, prompt reviews, technical reviews, troubleshooting, and technical mentoring.
Conduct AI-specific adversarial testing, including: Prompt injection, Jailbreak attempts, Data poisoning, Unsafe output, Unauthorized disclosure, Retrieval manipulation, Credential exposure.
Document findings, implement mitigations, perform retesting, and support release decisions.
Support ATO activities, security assessments, vulnerability remediation, privacy documentation, SBOM development, and compliance evidence.
Support AI applications that meet Section 508 and WCAG 2.1 Level A and AA requirements.
Serve as a Tier 2 and Tier 3 helpdesk escalation resource.
Be reachable within two hours for Tier 2 or Tier 3 tickets opened during business hours.
Provide after-hours escalation support for critical and high-severity AI incidents.
Support production issue triage, root-cause analysis, service restoration, user support, and continuous improvement.