Senior Agentic AI Developer

Job ID
2025-3321
# of Openings
1
Job Location
US-DC-Washington

Overview

***NO THIRD PARTIES***

 

MUST BE A US CITIZEN

TOP SECRET CLEARANCE REQUIRED

100% ONSITE - WASHINGTON, DC

 

Salary is dependent upon the candidate's skills, experience, education and certifications. 

Salary Range: 

High end of salary range is considered for those who exceed qualifications in one or more required skill areas. 

 

Soft Tech offers competitive BENEFITS in the areas of: Medical, Dental, Vision, 401K, Short Term Disability, Long Term Disability, Life Insurance, PTO, and Paid Holidays

 

Soft Tech Consulting is seeking a highly motivated, self-directed individual to fill the role of a Senior Cloud Database Administrator for one of our clients.  The designated candidate for this position will join a collaborative team environment, providing support and expertise necessary toward the successful delivery of the project deliverables, and provide effective communication to all levels of staff for reporting status/progress.

Responsibilities

  • Design and implement agentic AI workflows using AutoGen, CrewAI, LangChain, or an Agent Development Kit (ADK).
  • Integrate Model Context Protocol (MCP) for seamless AI-system communication.
  • Develop, test, and refine advanced prompts using systematic engineering methodologies.
  • Build full-stack generative AI applications on AWS Bedrock.
  • Develop Python backend services and React front-end interfaces.
  • Design scalable cloud architectures leveraging AWS services.

Qualifications

  • Bachelor's or Master’s in Computer Science or related field.
  • 8+ years professional technical/software development experience.
  • 1-2 years hands-on AI/ML development experience.
  • 1+ year production experience with AWS Bedrock and cloud services.
  • 2+ years Python programming experience.
  • 2+ years Front-end development experience with React.
  • Experience with additional frameworks (Semantic Kernel, Haystack, LlamaIndex).
  • Knowledge of other cloud platforms (Azure OpenAI, Google Vertex AI).
  • MLOps practices and containerization (Docker, Kubernetes).
  • Open-source AI contributions.
  • Must be able to communicate effectively both verbal and written.
  • Must be able to work independently, with minimal oversight.

Required Technical Skills:

  • Agentic AI Frameworks:
    • AutoGen, CrewAI, LangChain, or ADK
    • Model Context Protocol (MCP) implementation
    • Retrieval-Augmented Generation (RAG) architectures
  • GenAI & Prompt Engineering:
    • Advanced prompt design and optimization
    • Systematic testing and evaluation methodologies
    • Deep understanding of LLMs and their capabilities

Required Programming Skills:

  • Python: Advanced proficiency including async programming, testing frameworks.
  • React: Component development, state management, hooks.
  • RESTful API design and implementation.

Required AWS & Cloud Skills:

  • Amazon Bedrock (Claude, Titan, Llama models).
  • Lambda, ECS, S3, DynamoDB.
  • CloudWatch, IAM, security best practices.
  • Infrastructure as Code (CloudFormation/CDK).

Required Development Practices:

  • Agile/Scrum, test-driven development.
  • CI/CD pipeline implementation.
  • Code review and documentation.

About Us

Soft Tech Consulting, Inc. is a woman and minority-owned small business headquartered in Chantilly, VA. With contracts in both the public and private sectors in the DC metro area and across the country, Soft Tech is an organization made up of highly successful and talented Information Technology professionals offering enterprise class solutions for any size organization at great value. Soft Tech’s mission is to help government organizations design, implement, and maintain mission critical Information Technology solutions. By focusing jointly on our employees and our customers, we are able to achieve our mission by providing each and every one of our customers with continuous quality customer support.

 

Soft Tech Consulting, Inc is an Equal Opportunity Employer.

Options

Sorry the Share function is not working properly at this moment. Please refresh the page and try again later.
Share on your newsfeed