Generative AI / ML Engineer – Construction Scheduling

Qualifications

• We’re seeking a Generative AI / ML Engineer with 5+ years of experience in AI/ML and LLM-based systems
• 5+ years experience in AI/ML, ideally with LLM or generative AI projects
• Hands-on with foundation models (OpenAI, Claude, LLaMA, or similar)
• Experience with RAG, fine-tuning, prompt engineering, or LoRA
• Strong programming skills in Python; experience with APIs, FastAPI, or similar
• Familiarity with vector databases, data pipelines, and model evaluation/testing techniques
• Ability to translate domain knowledge into machine-readable formats
• Excellent written and verbal communication
• Fast-paced, startup environment
• Suppose you need to teach an AI model to understand construction schedules and real-world exceptions
• Which programming languages, frameworks, and deployment tools (FastAPI, Docker, Kubernetes, etc.)
• Describe a project where you fine-tuned or deployed a foundation model (GPT, Claude, LLaMA, etc.) for a specific domain

Benefits

• Hybrid: some onsite work in NYC, mostly remote
• Pay: $80,000.00 - $120,929.93 per year

Responsibilities

• You will design, build, and productionize AI models that understand construction schedules, specifications, and real-world exceptions
• You will work closely with domain experts (senior schedulers) to translate industry knowledge into machine-readable rules, RAG pipelines, and fine-tuned AI models
• Your work will power both internal prototypes and scalable applications for project planning, claims analysis, and scheduling optimization
• AI Model Development: Integrate foundation models (GPT-5, Claude, LLaMA) with scheduling and project data
• Build pipelines for prompt engineering, retrieval-augmented generation (RAG), and LoRA fine-tuning
• Data & Labeling: Collaborate with schedulers and data engineers to structure historical schedules, reports, and drawings for training and validation
• System Architecture: Plan AI workflows and interfaces for eventual API and web application integration
• Ensure models are secure, efficient, and auditable
• Testing & Evaluation: Define model validation pipelines with real-world examples
• Track accuracy, completeness, and sequence compliance
• Deployment Readiness: Prepare AI solutions for integration with backend APIs, front-end dashboards, and scheduling software
• Collaboration: Work with domain experts, product managers, backend/front-end developers to align technical solutions with business objectives
• Continuous Improvement: Research new techniques in generative AI, RAG, and LLM orchestration to enhance system performance
• What You’ll Do Day-to-Day
• Convert scheduler expertise into prompts, rules, and fine-tuning strategies
• Design data pipelines for ingesting schedules, drawings, and reports
• Build and test AI models with domain-specific benchmarks
• Collaborate with the team to prepare AI systems for integration into APIs and web apps
• Maintain AI performance tracking and provide recommendations for improvement
• Opportunity to shape the AI roadmap in construction scheduling and project management

Full Description

About ProjeCS

ProjeCS LLC is building AI-powered tools to transform construction project management.

We combine domain expertise in scheduling, risk management, and claims with cutting-edge AI to deliver intelligent copilot systems for project teams.

Our goal: make construction planning faster, more accurate, and actionable in the real world.

Job Summary

We’re seeking a Generative AI / ML Engineer with 5+ years of experience in AI/ML and LLM-based systems.

You will design, build, and productionize AI models that understand construction schedules, specifications, and real-world exceptions.

You will work closely with domain experts (senior schedulers) to translate industry knowledge into machine-readable rules, RAG pipelines, and fine-tuned AI models.

Your work will power both internal prototypes and scalable applications for project planning, claims analysis, and scheduling optimization.

Key Responsibilities

• AI Model Development: Integrate foundation models (GPT-5, Claude, LLaMA) with scheduling and project data.

Build pipelines for prompt engineering, retrieval-augmented generation (RAG), and LoRA fine-tuning.

• Data & Labeling: Collaborate with schedulers and data engineers to structure historical schedules, reports, and drawings for training and validation.

• System Architecture: Plan AI workflows and interfaces for eventual API and web application integration. Ensure models are secure, efficient, and auditable.

• Testing & Evaluation: Define model validation pipelines with real-world examples. Track accuracy, completeness, and sequence compliance.

• Deployment Readiness: Prepare AI solutions for integration with backend APIs, front-end dashboards, and scheduling software.

• Collaboration: Work with domain experts, product managers, backend/front-end developers to align technical solutions with business objectives.

• Continuous Improvement: Research new techniques in generative AI, RAG, and LLM orchestration to enhance system performance.

Required Qualifications

• 5+ years experience in AI/ML, ideally with LLM or generative AI projects.

• Hands-on with foundation models (OpenAI, Claude, LLaMA, or similar).

• Experience with RAG, fine-tuning, prompt engineering, or LoRA.

• Strong programming skills in Python; experience with APIs, FastAPI, or similar.

• Familiarity with vector databases, data pipelines, and model evaluation/testing techniques.

• Ability to translate domain knowledge into machine-readable formats.

• Excellent written and verbal communication.

Preferred Qualifications

• Multi-agent or agentic workflows experience.

• Production deployment experience (Docker, Kubernetes, cloud).

• Knowledge of construction project management, scheduling, or claims workflows.

• Experience working with structured/unstructured data and real-world exceptions.

• Familiarity with observability, model governance, and security in AI systems.

What You’ll Do Day-to-Day

• Convert scheduler expertise into prompts, rules, and fine-tuning strategies.

• Design data pipelines for ingesting schedules, drawings, and reports.

• Build and test AI models with domain-specific benchmarks.

• Collaborate with the team to prepare AI systems for integration into APIs and web apps.

• Maintain AI performance tracking and provide recommendations for improvement.

Work Environment

• Hybrid: some onsite work in NYC, mostly remote.

• Fast-paced, startup environment.

• Opportunity to shape the AI roadmap in construction scheduling and project management.

Job Types: Full-time, Part-time

Pay: $80,000.00 - $120,929.93 per year

Application Question(s):

• Suppose you need to teach an AI model to understand construction schedules and real-world exceptions. How would you work with domain experts to structure this knowledge for the model?

• Which programming languages, frameworks, and deployment tools (FastAPI, Docker, Kubernetes, etc.) have you used to productionize AI models? Please give an example.

• How do you validate that an AI/ML model is performing correctly in a real-world setting? What metrics or testing strategies have you used?

• Describe a project where you fine-tuned or deployed a foundation model (GPT, Claude, LLaMA, etc.) for a specific domain. What approach did you take, and what were the results?

• Have you implemented retrieval-augmented generation (RAG) pipelines?

What data sources did you use, and how did you ensure model outputs were accurate and traceable?

Work Location: In person