US Citizenship is required

In person interview is required.

The Judge Group is seeking an AI/ML Solutions Architect (Agentic Systems) to design and deliver secure, scalable, and cost-effective cloud-native AI solutions for federal clients. You will bridge complex mission needs and modern technology by owning end-to-end architectures and leading hands-on implementation—especially for agentic AI systems, RAG-based applications, and production-grade ML pipelines.

This role blends technical vision + practical delivery leadership: selecting tools, defining architectures, establishing engineering standards, and guiding implementation through prototypes, code reviews, and reference solutions.

What You Will Do:

• Design and implement agentic AI systems that enable autonomous decision-making, workflow orchestration, and mission process optimization—with appropriate guardrails and human oversight.

• Develop Generative AI applications for summarization, extraction, predictive insights, and conversational interfaces.

• Build and maintain scalable data pipelines integrating structured + unstructured data to support analytics and AI workloads.

• Apply advanced statistical and machine learning techniques to decision support and policy/program evaluation.

• Lead AI initiatives spanning:

• Retrieval-Augmented Generation (RAG) and evaluation

• Re-ranking strategies and retrieval quality optimization

• Prompt engineering, safety patterns, and defensive design

• Knowledge graph integration and graph-enhanced retrieval

• AI chatbots and conversational agents

• Fine-tune embeddings and LLMs (when appropriate) to improve domain performance, accuracy, robustness, and retrieval quality.

• Build entity graphs using entity resolution (matching, deduplication, linking, relationship discovery) to enable graph analytics and enhanced retrieval.

• Collaborate across engineering, security, and stakeholders to prototype rapidly, iterate responsibly, and deliver mission-ready outcomes.

• Lead deployment in AWS-first cloud environments, leveraging Infrastructure-as-Code, DevOps/DevSecOps, and operational excellence patterns.

• You will own and drive the technical foundation and delivery rigor for mission AI solutions:

• End-to-end solution architecture: system boundaries, trust zones, data flows, integrations/APIs, security controls, observability, and cost models.

• Tooling and platform selection: LLMs, embeddings, vector stores, orchestration frameworks, graph technologies, data platforms—documenting tradeoffs and decisions.

• Engineering and delivery standards: secure SDLC, CI/CD quality gates, automated testing, code review practices, evaluation harnesses, and production readiness checklists.

• Hands-on technical leadership: prototypes, reference implementations, PR reviews, mentoring, and architecture governance to ensure delivery quality.

What You Will Need:

• Must be able to OBTAIN and MAINTAIN a Federal or DoD "PUBLIC TRUST"; candidates must obtain approved adjudication of their PUBLIC TRUST prior to onboarding. Candidates with an ACTIVE PUBLIC TRUST or SUITABILITY are preferred.

• Bachelor’s degree in Engineering, IT, Computer Science, or related field (or equivalent experience).

• Minimum EIGHT (8) years in solutions architecture, software engineering, data engineering, and/or applied ML with a track record of delivering production systems. A Master’s degree may be substituted for up to 2 years of relevant professional experience.

• Strong Python proficiency and strong SQL skills (data modeling, query optimization).

• Experience designing and delivering cloud-based AI/ML solutions end-to-end (ingestion → modeling → deployment → monitoring) in secure environments.

• Hands-on experience with AI application frameworks such as LangChain, Haystack, crewAI, or similar.

• Strong knowledge of core Python ML/data libraries: NumPy, Pandas, Scikit-learn, NLTK, OpenCV.

• Familiarity with deep learning frameworks such as PyTorch or TensorFlow.

• Experience with search technologies such as Elasticsearch or OpenSearch.

• Experience with relational databases (e.g., PostgreSQL, Oracle) and in-memory analytics engines (e.g., DuckDB).

• Experience using cloud SDKs (e.g., Boto3) and building reliable integrations with cloud services.

• Familiarity with agentic AI frameworks such as AWS Strands Agents, PydanticAI, and related orchestration patterns.

• Advanced prompt engineering skills for complex tasks beyond code generation (reasoning workflows, guardrails, evaluations).

• Experience with asynchronous Python development (asyncio patterns, concurrency, reliability).

• Experience with MCP servers and tool-calling within agentic workflows (tool governance, reliability, and security considerations).

By providing your phone number, you consent to: (1) receive automated text messages and calls from the Judge Group, Inc. and its affiliates (collectively “Judge”) to such phone number regarding job opportunities, your job application, and for other related purposes. Message & data rates apply and message frequency may vary. Consistent with Judge's Privacy Policy, information obtained from your consent will not be shared with third parties for marketing/promotional purposes. Reply STOP to opt out of receiving telephone calls and text messages from Judge and HELP for help.