Lead AI Platform Engineer

Lead AI Platfrom EngineerAI-Native Software Development & Production LLM SystemsLocation: Los Angeles Metropolitan AreaEmployment Type: Full-time, on-siteAbout 190 Intel190 Intelligence is a protective intelligence and threat management company serving private clients, security organizations, and law enforcement agencies.Founded by veterans of national security, intelligence, and protective operations, 190 Intel combines proprietary technology with experienced intelligence professionals and field operations. Powered by VENONA, our unified intelligence platform, we monitor millions of digital indicators, track thousands of threat actors, and use AI to analyze and triage signals from the digital and physical worlds.Our work spans threat identification, attribution, assessment, investigation, and ongoing threat management. Our California-based Central Operations team provides 24/7 monitoring, situational awareness, and crisis support in coordination with clients and law enforcement partners.About the role:190 Intel is seeking a Lead AI Platform Engineer to lead the architecture and development of the technology at the core of our VENONA platfrom.This is not a conventional software engineering role. AI is central to both our product and the way we build it.You will define the platform architecture, direct coding agents, critically review the software they produce, and recognize when plausible-looking code is incomplete, insecure, or incorrect. You will remain accountable for ensuring that the final system is secure, reliable, maintainable, and ready for production.You will also own the AI capabilities within our product, including LLM pipelines, prompt and workflow design, model selection, batch processing, evaluation, accuracy, latency, and cost.This is a hands-on technical leadership role for an experienced engineer who already works in an AI-native way and wants to apply that approach to challenges with meaningful real-world consequences.ResponsibilitiesDefine and evolve the architecture of the VENONA platform.Use and direct coding agents to accelerate software design, development, testing, documentation, and troubleshooting.Critically review, test, and validate agent-generated code rather than accepting plausible-looking output at face value.Remain accountable for the quality, security, maintainability, and production readiness of the completed software.Design, build, and operate production-grade applications, services, APIs, and data-processing workflows.Own the development and performance of the LLM capabilities within the product.Design and optimize LLM pipelines, prompts, workflows, model selection, and batch-processing systems.Build evaluation processes that measure model accuracy, consistency, latency, reliability, and cost.Establish appropriate testing, monitoring, observability, and quality-control practices for both software and model outputs.Identify and address failure modes across application code, data pipelines, model behavior, and infrastructure.Work closely with product leaders, intelligence analysts, investigators, and operations teams to translate real-world workflows into effective technical solutions.Establish engineering standards and help shape how 190 Intel develops software in an AI-native environment.Required qualifications:Strong software engineering foundations, including substantial experience with Node.js, TypeScript, SQL, APIs, and system design.Experience architecting, building, and operating production-grade software systems.Meaningful experience using coding agents to build software that has reached production.Ability to direct coding agents effectively, review their output critically, and independently verify correctness.Hands-on experience building, deploying, and evaluating production LLM applications.Practical experience with LLM pipelines, prompt and workflow design, model selection, and model integration.Experience designing evaluation methods for model accuracy, consistency, latency, reliability, and cost.Strong understanding of databases, data modeling, application architecture, and backend development.Ability to troubleshoot complex issues across code, models, data, APIs, and infrastructure.Strong analytical ability and technical judgment, particularly when evaluating outputs that appear credible but may be incorrect.Experience making architectural decisions and taking ownership from initial design through production operation.Strong written and verbal communication skills, including the ability to work effectively with technical and nontechnical stakeholders.Ability to work on-site in the Los Angeles area.Preferred qualifications:Experience building AI-enabled products that analyze large volumes of unstructured or multimodal information.Experience processing social media, text, imagery, video, or other complex data sources.Familiarity with model evaluation frameworks, tracing, observability, and production monitoring.Experience with batch-processing systems and high-volume data pipelines.Experience designing secure systems that handle sensitive or mission-critical information.Familiarity with cloud infrastructure, deployment automation, application security, and scalable system operations.Knowledge of threat intelligence, investigations, public safety, protective operations, or related fields.Experience leading technical initiatives, mentoring engineers, or helping build an early-stage engineering organization.Bachelor’s or advanced degree in computer science, engineering, or a related discipline; Successful role outcomes:Coding agents materially improve engineering speed without compromising quality, security, or maintainability.Agent-generated code is subject to disciplined review, testing, and independent validation before reaching production.LLM capabilities deliver accurate, consistent, measurable, and cost-effective results.Intelligence analysts and investigators can use the platform to assess complex information and reach actionable conclusions more efficiently.The platform remains secure, reliable, and observable as its capabilities, data volumes, and customer usage expand.190 Intel develops a repeatable AI-native engineering model that can scale as the team grows.