1 month ago
Arlington, VA, USAMid Level
Base Salary
$125k - $145k/yr
Responsibilities
- Design agent definitions, context, tool allowlists, model selection, guardrails, and multi-agent architectures.
- Build shared, versioned agent infrastructure supporting interactive, scheduled, CI-triggered, event-driven, and unattended workloads.
- Define autonomy boundaries, credential scoping, cost and runtime limits, escalation paths, observability, and fail-open or fail-closed behavior.
- Create rigorous evaluations, validation harnesses, A/B trials, ablation studies, blinded LLM judging, uncertainty reporting, and contamination controls.
- Build layered redaction, egress filtering, and data-layer enforcement for agents operating near controlled information.
- Integrate agent tooling with requirements management, issue tracking, documentation, source control, diagramming, unstructured data, and enterprise identity systems.
- Operate production services with uptime expectations, incident response, and root-cause analysis.
- Develop onboarding, curricula, setup automation, self-diagnosing tooling, and scalable enablement for technical and non-software audiences.
- Review contributor work, maintain platform standards, grow the contributor base, and translate user pain into organizational priorities.
Requirements
- At least 3 years of experience building software systems, with meaningful recent depth in LLM-based or agentic systems.
- Fluency with tool or function calling, MCP or an equivalent tool protocol, context management, retrieval, multi-agent orchestration, and policy-enforcement hooks or middleware.
- Practical judgment regarding model selection and cost, latency, and quality tradeoffs.
- Experience evaluating AI system quality rigorously and analyzing unstructured data.
- Experience running non-interactive production workloads such as CI pipelines, scheduled jobs, or event-driven triggers.
- Strong Python skills and comfort with shell environments, CI, and Linux service operation.
- Experience building and maintaining production applications.
- Working knowledge of enterprise identity and authorization, including OAuth2/OIDC/SAML, tokens, scopes, and least privilege.
- Excellent written and oral communication skills.
- Preferred: statistics, classical machine learning, deep learning, LLM fine-tuning with DPO, QLoRA, or GRPO, open-source agent-harness contributions, knowledge graphs, ontologies, retrieval architecture, engineering productivity metrics, internal developer platforms, systems or safety engineering, verification and validation, regulated or classified AI environments, human-in-the-loop review gates, agent-evaluation publications or contributions, and robotics, autonomy, or safety-critical software experience.
- May require eligibility to obtain and maintain a U.S. security clearance.
Benefits
- Base salary range of $125,000-$145,000, with equity included in most full-time high-demand roles.
- Three premium healthcare plan options, including an HSA-eligible plan, with Forterra covering 80% of premiums for the employee and dependents.
- Company-paid basic life/AD&D and short- and long-term disability insurance, with additional life insurance available for purchase.
- Company holiday calendar with a December winter break and 20 accrued PTO days per year.
- At least 7 weeks of fully paid parental leave for birth or adoption.
- $9,000 annual tuition reimbursement or professional development stipend.
- 401(k) plan with traditional, Roth, and after-tax deferral options and company matching up to 4%.
- Fully stocked beverage refrigerators.
- Security-clearance eligibility may be required.
Categories
About Forterra
Forterra builds autonomous mission systems for defense and industrial ground operations, including self-driving kits for land vehicles and coordinated robotic swarms. It develops mission-critical hardware and software for autonomy in GPS-denied and contested environments, supporting military logistics, convoying, and yard/drayage use cases. Founded in 2002 and headquartered in Clarksburg, Maryland, the privately held company serves U.S. and allied government programs alongside commercial fleet operators.