AI Forward Deployed Engineer

CGS; Maryland; On-site; USD 150,000 - 250,000; Full time

Job summary

CGS is seeking Forward Deployed Engineers to build and deploy AI systems for government missions, at the intersection of software, applied AI, and real-world constraints. You will own projects from concept to production and collaborate with government stakeholders.

This early career MT role offers autonomy and impact, with on-site work in the DC region and opportunities to work with federal partners. Join as a founder-like engineer; bring strong Python skills, experience shipping LLMs, and a

Qualifications

• 1–4 years of professional engineering experience with genuine production ownership.
• Strong Python skills and end-to-end software engineering ability.
• Hands-on experience shipping LLM and agent systems to production.
• Experience with RAG, tool calling, multi-agent systems, and context management.
• Experience building data pipelines and ML infrastructure.
• Strong fundamentals and ability to communicate with government stakeholders.

Responsibilities

• Build and deploy LLM-powered agent systems for real-world government applications.
• Develop production systems using RAG, tool calling, multi-agent architectures, and advanced context management.
• Build data pipelines and ML infrastructure supporting AI systems.
• Take features and systems from concept through production.
• Work directly with government stakeholders to understand problems and translate them into technical solutions.
• Communicate complex technical concepts clearly to non-technical audiences.
• Iterate quickly based on real-world feedback and changing requirements.

Skills

Python

LLM systems

Agent systems

RAG

Tool calling

Multi-agent

Context management

Data pipelines

ML infrastructure

Ownership

Government stakeholder communication

Job description

$150,000–$250,000 base + 0.5–1% equity | In-office 5 days/week | 7 openings

Level: Member of Technical Staff

Experience: 1–4 years

Start: Candidates should be available to start soon

About the Company

This AI company is building technology to systematically identify and combat fraud, waste, and abuse in US government spending. Its agentic platform combines public and government-restricted data including federal spending and contracting databases, Medicare and Medicaid claims and reasons across that information to identify specific cases of potential fraud.

The platform doesn't just surface anomalies. It provides the evidence chain auditors need to investigate and take action. The company is already in active discussions with federal health agencies, defense, homeland security, and state governments. With significant contracts within reach, the biggest constraint is now engineering capacity. The company is building its first cohort of Forward Deployed Engineers to work directly with government partners.

About the Role

We're looking for early-career engineers who already operate like founders. As a Forward Deployed Engineer, you'll sit at the intersection of software engineering, applied AI, and real-world government missions. You'll build and deploy AI systems that solve high-impact problems, working directly with government stakeholders and taking ownership from initial concept through production.

This is a Member of Technical Staff role with significant autonomy and responsibility.

What You'll Own

• Build and deploy LLM-powered agent systems for real-world government applications
• Develop production systems using RAG, tool calling, multi-agent architectures, and advanced context management
• Build data pipelines and ML infrastructure supporting AI systems
• Take features and systems from concept through production
• Work directly with government stakeholders to understand problems and translate them into technical solutions
• Communicate complex technical concepts clearly to non-technical audiences
• Iterate quickly based on real-world feedback and changing requirements

What We're Looking For

1–4 years of professional engineering experience with genuine production ownership.

We're looking for engineers with:

• Strong Python skills and end-to-end software engineering ability
• Hands-on experience shipping LLM and agent systems to production
• Experience with RAG, tool calling, multi-agent systems, and context management
• Experience building data pipelines and ML infrastructure
• Strong engineering fundamentals, not just AI coding-tool proficiency
• The ability to represent the company confidently with government stakeholders
• A strong sense of ownership and a bias toward action

Strong Backgrounds

Particularly relevant experience includes:

• Founder or founding engineer at an AI startup
• Forward-deployed or applied engineering at companies such as Palantir, Anduril, Shield AI, Vannevar Labs, Scale AI Federal, or Applied Intuition
• Applied engineering at agentic AI companies such as Sierra, Decagon, Glean, Cresta, Perplexity, Cursor, or Cognition
• Experience at US government technology organizations such as USDS or 18F
• Big Tech experience combined with clear evidence of zero-to-one ownership
• Experience from startup accelerators or fellowships such as Neo, Z Fellows, South Park Commons, or Pioneer

Nice to Have

• Active US security clearance, Secret, Top Secret, or TS/SCI
• Experience with Medicare, Medicaid, or healthcare fraud
• Fine-tuning and model evaluation experience
• Reinforcement learning or simulation environments
• Existing connections to Washington, DC or Northern Virginia

Government & Location Requirements

There are two teams within the Forward Deployed Engineering group.

1. Defense & Homeland Security

• US citizenship is required
• Work on-site at the Pentagon approximately 2–3 days per week

2. Federal Health & State/Local Government

• Exceptional O-1 candidates may be considered
• Work directly with federal health agencies and state/local governments

For all candidates:

• Must be based in Washington, DC or willing to relocate quickly
• Must work in person 5 days per week
• Must be available to start soon

Who This Role Is Not For

This opportunity is unlikely to be a fit for:

• Candidates with internship-only experience
• Engineers without strong software fundamentals
• Candidates who rely heavily on AI coding tools without understanding the underlying systems
• Big-firm consulting backgrounds without meaningful product or engineering ownership
• Pure ML researchers without production engineering experience
• Engineers who prefer to work independently without customer or stakeholder interaction
• Candidates unwilling to relocate to DC or work in person
• Candidates who are uncomfortable working on US government applications

Interview Process

The process is designed to assess both technical depth and the ability to operate effectively with customers and stakeholders.

You'll be evaluated on your engineering fundamentals, production AI experience, problem-solving ability, communication, and ownership.

This is an opportunity to join an early FDE team, work directly on high-impact government problems, and build production AI systems where the results have real-world consequences.