Sr. AI Researcher, Agentic Systems

AItoolnavio; United States; Remote; USD 184,000 - 265,000; Full time

Benefits offered by this job

Equity awards

Medical/dental/vision coverage

Flexible PTO

401(k) matching

Parental leave

Home office stipend

Employee referral program

Annual company retreats

Job summary

AItoolnavio is seeking a Senior AI Researcher specializing in agentic systems to build intelligent agents that reason across multimodal construction data, including images, BIM, drawings and specs. The role blends research and production engineering in a fast-paced, remote US setting.

You will design agent architectures, retrieval layers, and evaluation frameworks, selecting models to balance latency, cost and accuracy while ensuring traceable outputs with verifiable evidence.

Qualifications

• Master's or Ph.D. in a relevant field, or equivalent professional experience.
• At least 5 years of experience building machine learning systems, including production of LLM- or VLM-based systems.
• Ability to design evaluation frameworks from scratch where ground truth is scarce or costly.
• Strong knowledge of multimodal AI, including vision-language models and retrieval.
• Strong software engineering skills, especially Python, asynchronous orchestration, tracing and cost instrumentation.
• Ability to build custom testing and evaluation harnesses for AI systems.
• Familiarity with spatial data concepts like coordinate frames and camera poses.
• Good technical judgment on automation readiness and reliability.
• Excellent collaboration and ability to translate research into production-ready solutions.

Responsibilities

• Design and develop agentic systems capable of planning and executing multi-step tasks across imagery, spatial geometry, and technical documents.
• Build and maintain the context and retrieval layer enabling access to evidence from spatially and temporally indexed records.
• Develop tools/interfaces for locating components, comparing site captures, and querying BIM elements.
• Design evaluation frameworks for complex tasks with limited ground truth.
• Use evaluation results to determine readiness and guide improvements before production.
• Select appropriate models balancing frontier APIs, open-weight models, fine-tuned models and classic CV methods with cost/latency/accuracy.
• Ensure outputs are traceable to supporting evidence in images, drawings, or specs.
• Prototype new approaches rapidly and collaborate with platform engineers to productionize successful systems.
• Monitor and optimize deployed systems for reliability, performance and cost efficiency.

Skills

Multimodal AI

Production ML systems

Evaluation frameworks

Research collaboration

Education

Master's or Ph.D. in a relevant field

Tools

Python

Retrieval systems

Cost instrumentation

Asynchronous orchestration

Job description

This position is listed on behalf of a partner company, who manages all applications and next steps. Our partner is looking for a Sr. AI Researcher, Agentic Systems based in the United States.

Join an innovative team developing advanced AI systems that transform how complex construction projects are understood and managed. As a Senior AI Researcher specializing in agentic systems, you will build intelligent agents capable of reasoning across multimodal construction data, including 360° imagery, point clouds, BIM models, drawings, and technical specifications. Your work will bridge applied research and production engineering, turning cutting‑edge AI capabilities into reliable, actionable insights. You will design agent architectures, retrieval systems, evaluation frameworks, and tools that help models answer real‑world questions with verifiable evidence. Working closely with computer vision and platform engineering teams, you will shape the next generation of spatial AI applications. This remote opportunity is ideal for an experienced ML professional who thrives on complex challenges, rapid experimentation, and measurable real‑world impact.

Accountabilities:

• Design and develop agentic systems capable of planning and executing multi-step tasks across imagery, spatial geometry, and technical documents.
• Build and maintain the context and retrieval layer, enabling agents to access relevant evidence from spatially and temporally indexed construction records.
• Develop the tools and interfaces agents use to locate components, compare site captures, inspect drawing regions, and query BIM elements.
• Design evaluation frameworks for complex tasks where ground truth is limited, expensive to obtain, or subject to interpretation.
• Use evaluation results to determine system readiness, identify limitations, and guide improvements before production deployment.
• Select the most appropriate model for each task, balancing frontier-model APIs, open‑weight models, fine‑tuned smaller models, and classical computer vision techniques against cost, latency, and accuracy requirements.
• Ensure generated outputs are traceable by connecting claims to their supporting evidence in images, drawing sheets, or specification sections.
• Prototype new approaches rapidly, discontinue ineffective solutions, and collaborate with platform engineers to bring successful systems into production.
• Monitor and optimize deployed systems to maintain reliability, performance, and cost efficiency at scale.

Requirements:

The ideal candidate combines advanced machine learning expertise, strong software engineering capabilities, and sound research judgment. You should be comfortable building production‑grade AI systems, evaluating their limitations rigorously, and working with multimodal and spatial data.

• Master's or Ph.D. in a relevant field, or equivalent professional experience.
• At least 5 years of experience building machine learning systems, including experience taking at least one LLM‑ or VLM‑based system into production and maintaining it over time.
• Demonstrated ability to design evaluation frameworks from scratch when reliable ground truth is sparse, costly, or contested.
• Strong knowledge of multimodal AI, including vision‑language models, embeddings, retrieval systems, context engineering, and fine‑tuning when justified by performance or cost.
• Strong software engineering skills, particularly in Python, asynchronous orchestration, tracing, and cost instrumentation.
• Ability to build custom testing and evaluation harnesses for AI systems.
• Familiarity with spatial data concepts, including coordinate frames, camera poses, and projections.
• Strong technical judgment in determining which tasks are appropriate for automation and when a system is not yet reliable enough to provide an answer.
• Ability to collaborate effectively with research and engineering teams, communicate findings clearly, and translate experimental results into production‑ready solutions.

Preferred qualifications:

• Experience with post‑training techniques or reinforcement learning applied to agent trajectories.
• Familiarity with document understanding for technical drawings, specifications, and other structured PDF documents.
• Experience working with additional sensor data, including mobile capture, wearable devices, or field audio.
• Background in construction, robotics, geospatial technology, or another domain where AI models interact with the physical world.

Benefits:

• Base salary: $184,000–$265,000 annually, depending on location, skills, experience, training, qualifications, and other business considerations.
• Equity awards as part of the total rewards package.
• Medical, dental, and vision coverage with 100% of premiums paid by the employer.
• Flexible paid time off.
• 401(k) retirement plan with company matching.
• Paid parental leave.
• Home office stipend.
• Employee referral program.
• Annual company retreats and team gatherings.
• Region-specific health and wellness benefits.
• Remote work arrangement within the United States.