Top Tech Jobs & Startup Jobs in Chicago, IL

Evaluate and improve AI-generated biomedical engineering content for technical accuracy, logical soundness, completeness, safety, and regulatory awareness. Develop realistic biomedical scenarios and reasoning tasks, review prompts and step-by-step solutions, identify errors and unsupported claims, and provide structured feedback. Assess model performance across medical devices, biomechanics, bioinstrumentation, signals, biomaterials, imaging, verification, validation, and risk management while promoting responsible communication of assumptions, limitations, and patient-safety considerations.

Use CATIA to create and edit 3D mechanical parts, assemblies, surfaces, and engineering drawings. Explain and document CAD workflows step by step, review technical tasks, provide feedback supporting AI model improvement, and ensure design accuracy and quality.

Use GIMP to complete image editing, photo retouching, and 2D graphic design tasks. Record and explain editing workflows, manage layers and visual elements, document processes, and review tasks to provide feedback for AI model improvement while maintaining accuracy and quality.

Design and execute AI security evaluations focused on ethical jailbreaks, LLM red teaming, prompt injection, and tool-use abuse. Develop adversarial elicitation strategies, regression test suites, and evaluation frameworks to identify vulnerabilities in AI models. Collaborate with technical stakeholders to translate findings into safety improvements, and document methodologies and results for technical and non-technical audiences.

Conduct equity research analysis for AI training datasets, including financial modeling, earnings analysis, valuation, peer benchmarking, sector research, and investment thesis development. Review company models, assess revenue, margins, cash flow, and valuation drivers, produce concise research summaries and event updates, identify industry trends and catalysts, and provide quality feedback to project stakeholders.

Create reinforcement learning environments to train and evaluate AI systems using Model Context Protocol tools. Responsibilities include designing deterministic verification and reference solutions, debugging software, implementing features, refactoring codebases, optimizing performance, and assessing AI agents’ software engineering and MCP tool-use abilities.

Create reinforcement learning environments to evaluate AI systems using Model Context Protocol tools. Responsibilities include designing reproducible environments, deterministic verification, and reference solutions while fixing bugs, implementing features, refactoring code, and optimizing software performance across complex codebases.

Create and revise production-ready parametric 3D CAD models, mechanical parts, and assemblies using FreeCAD or comparable software. Produce engineering drawings with GD&T and tolerance stack-ups, design for sheet metal, CNC, plastics, and injection molding, and validate CAD models for manufacturability. Prepare technical documentation and assembly instructions for AI model training, incorporate reviewer feedback, and ensure accurate manufacturing handoff.

Evaluate AI model responses to complex business strategy and consulting problems. Develop high-quality prompts, model answers, and detailed feedback; apply structured consulting frameworks to assess reasoning quality, identify gaps, and help improve model performance. Collaborate with the project team as evaluation guidelines evolve. This is a remote contract role requiring strong consulting experience, business analysis, structured problem-solving, and excellent written English.

Conduct coding-heavy machine learning research involving model development, training pipelines, inference systems, numerical methods, performance optimization, and validation. Build reproducible workflows in Python and command-line tools, implement model and data pipeline components, debug system failures, and verify correctness and performance. The contract is fully remote, flexible, and approximately 15 hours per week, with output-based compensation.