Strategy Consultant - AI Training & Evaluation (MBB & Top-Tier Firms)

Role overview Toloka AI is hiring a Strategy Consultant to focus on AI training and evaluation. This remote role is based in Stuttgart, Baden-Württemberg, Germany. The position involves close collaboration with leading consulting firms, including MBB and other top-tier organizations, to strengthen AI-driven solutions. What you will do Work with teams from major consulting firms to design and improve AI strategies. Apply expertise in AI training and evaluation to help meet client goals. Shape methods that support clients in achieving results with AI technologies.

Role overview toloka-ai seeks a Senior Consultant with a deep background in artificial intelligence for freelance assignments. This role involves guiding AI projects, sharing expertise built at leading firms, and helping to develop solutions that align with client objectives. The consultant will play a key part in enhancing toloka-ai’s service offerings through direct project involvement. What you will do Advise on AI projects from initial concept through delivery Draw on experience from top-tier firms to address client needs Work closely with a remote team to design and improve AI solutions Contribute to the ongoing development of toloka-ai’s services through active project work Location This freelance position is fully remote. Candidates can be based in Stuttgart, Baden-Württemberg, Germany, or work from another location.

Toloka-AI seeks a Computer Science Specialist with advanced Python skills to join its remote team based in Stuttgart, Baden-Württemberg, Germany. This role centers on supporting AI-driven projects that serve a global audience. Key Responsibilities Design and improve artificial intelligence solutions using Python Collaborate with fellow specialists on projects with worldwide impact Tackle challenging technical problems in the AI domain Make meaningful contributions to projects with practical applications About Toloka-AI Toloka-AI develops products powered by artificial intelligence, aiming for broad global use and tangible results. The company emphasizes teamwork and a commitment to technical quality in every initiative.

Please submit your CV in English and indicate your English proficiency level. Mindrift connects skilled specialists with project-based AI work at leading technology companies. Assignments focus on testing, evaluating, and improving AI systems. This is a freelance, project-based role and does not offer permanent employment. Role Overview Mindrift is building a dataset to assess AI coding agents on real-world developer tasks. As a Freelance AI Evaluation Engineer, the work centers on designing detailed tasks and evaluation methods within realistic simulated environments. Main Responsibilities Create virtual companies from high-level plans, including codebases, infrastructure, and realistic context such as conversations, documentation, and tickets that reflect authentic development history. Develop and refine tasks based on different stages of the virtual company. This includes writing prompts, setting evaluation criteria, and ensuring tasks are solvable and assessments are fair. Design assignments for isolated environments that mimic a developer's workstation, using a Linux machine with development tools (terminal, CLI), MCP servers (repository, task tracker, messenger, documentation), and a real web application codebase. Build tests that accept all valid solutions and reject incorrect ones, aiming for a balanced level of strictness. Work with an AI agent to confirm that tests detect real issues, do not overlook errors, and validate correct solutions. Review code generated by agents, analyze why solutions succeed or fail, and invent edge cases and adversarial scenarios. Incorporate feedback from expert QA reviewers to improve your work and meet quality standards. Scope Clarifications This position does not include data labeling. This position does not cover prompt engineering. You will not be required to write code from scratch. The AI agent handles most coding; your focus is on guidance and evaluation. Much of the work involves direct collaboration with AI, as designing challenges for advanced models requires using those models in practice.

Role Overview Toloka AI is seeking a Chemistry and Python Specialist to join as a Freelance AI Training Consultant. This remote position is open to candidates based in Stuttgart, Baden-Württemberg, Germany. The role centers on developing and delivering training materials for learners interested in artificial intelligence. What You Will Do Create educational content that connects chemistry concepts with practical Python programming. Design and conduct training sessions for a community of aspiring AI professionals. Help learners apply theoretical knowledge to real-world AI challenges. What We’re Looking For Strong background in chemistry. Proficiency in Python programming. Experience developing or delivering technical training materials is a plus. Interest in artificial intelligence and education. Location and Engagement This is a freelance, remote role. Candidates should be based in Stuttgart, Baden-Württemberg, Germany.

Role Overview Toloka AI is looking for a Freelance Legal Consultant with expertise in US law to support the training of AI systems. This fully remote position is based in Stuttgart, Baden-Württemberg, Germany, and offers flexible working arrangements. What You Will Do Apply knowledge of US legal concepts to improve AI training processes Collaborate with technical teams to guide the development of AI applications Review and provide feedback on legal content used in AI models What We Look For Strong background in US law Experience interpreting and explaining legal concepts Comfort working independently in a remote setting

Toloka AI seeks a Freelance English Writer and AI Trainer to join its remote team. This contract position is open to candidates based in Stuttgart, Baden-Württemberg, Germany, but all work can be completed remotely. Role overview This role centers on supporting AI projects that require strong English writing and language abilities. The work involves both content creation and helping AI systems better understand natural language. What you will do Write and edit English-language content for use in AI training Contribute to projects aimed at improving AI's language understanding Collaborate with the Toloka AI team to achieve project objectives Location Remote work is available. The position is based in Stuttgart, Baden-Württemberg, Germany, but candidates may work from anywhere.

Toloka AI seeks a Freelance Junior Journalist to contribute to AI training projects. This position is remote, based in Stuttgart, Baden-Württemberg, Germany. The main focus is on creating and refining training data that supports the development of artificial intelligence systems. Key responsibilities Research topics in AI and technology to inform project work Write and edit content used in training data sets Curate materials that help AI systems better understand and interact with users Work collaboratively with a distributed team committed to improving AI capabilities Work environment This freelance, remote role involves working alongside colleagues from diverse backgrounds. The team shares a commitment to making AI more effective and accessible.

Toloka AI seeks a Freelance Mathematics Expert for remote contract work based in Stuttgart, Baden-Württemberg, Germany. This role centers on applying advanced mathematical concepts to support ongoing AI training projects and improve algorithm performance. Key responsibilities Use mathematical methods to develop and refine AI algorithms Contribute expertise to projects focused on advancing intelligent systems Collaborate with a distributed team of AI professionals Collaboration and work environment This position offers the chance to work remotely with colleagues from varied backgrounds. Sharing mathematical expertise is central to shaping new AI technologies as part of a global team.

Location: Remote , Stuttgart, Baden-Württemberg, GermanyCompany: Mindrift Project-Based Role at Mindrift Mindrift connects experienced specialists with project-based AI work for top technology companies. This role is not a long-term employment position but offers the chance to contribute expertise to focused AI initiatives. What You’ll Do Create realistic supply chain disruption scenarios based on hands-on manufacturing and procurement experience. Examples include supplier delays, changes in order quantities, logistics hurdles, or quality failures. Define expected outcomes and outline practical mitigation strategies for each scenario. Assess AI-generated responses and recommendations, checking for accuracy, completeness, and fit for real business use. Review outputs for relevance within ERP systems, especially Microsoft Dynamics 365, Coupa, Jaggaer, and Ariba (SAP). Contribute to structured data creation and validation, following provided guidelines and frameworks. What We Look For At least 4 years of experience in procurement, supply chain management, or purchasing, ideally in manufacturing. Solid understanding of procurement workflows: purchase orders, vendor management, inventory control, and production planning. Hands-on experience with ERP platforms such as SAP, Oracle, or Microsoft Dynamics 365. Ability to design and justify supply chain disruption scenarios and related mitigation strategies. Knowledge of disruption types: delays, shortages, quality issues, and logistics challenges. Familiarity with Incoterms and logistics management. Understanding of Bill of Materials (BOM) structures and production planning methods. Experience with supplier performance metrics, including On-Time In-Full, lead times, and quality scores. Strong analytical skills for evaluating AI outputs against business logic. Background in data validation, structured data entry, or annotation tasks. Clear written communication in English. How to Join Apply Pass qualifying assessments Join a project Complete assigned tasks Receive payment Project Commitment Active phases typically require about 10-20 hours per week, depending on project needs. This estimate may vary and does not guarantee a set number of hours. Compensation Contributors may earn up to $60 per hour, depending on expertise and project contributions.

Please submit your CV in English and include your English proficiency level. About Mindrift Mindrift connects skilled specialists with project-based AI work for major technology companies. The focus is on testing, evaluating, and improving AI systems. This is a project-based contract, not a permanent position. Role Overview As an Optical Engineer specializing in Python for AI training, work centers on developing and validating computational physics challenges that support AI research. Tasks may include: Designing original computational physics problems that mirror real research workflows Writing problems that require Python programming to solve, using tools such as Numpy, SciPy, and Sympy Ensuring challenges are computationally intensive and not solvable by hand within a reasonable timeframe Developing complex problems in mechanics, electromagnetism, thermodynamics, and quantum mechanics Basing problems on real research questions or practical applications in physics Validating solutions in Python with standard physics simulation libraries Documenting problem statements and providing verified solutions Who Should Apply This project suits optical engineers with strong Python skills who are interested in flexible, part-time work. The ideal candidate has: A degree in Physics (theoretical, experimental, computational) or a closely related field Proficiency in Python for numerical validation; experience with MATLAB, R, C, SQL, Numpy, Pandas, SciPy, Stata, or other programming languages is also welcome At least 2 years of professional experience in applied, research, or teaching roles Background in numerical simulation methods Experience designing problems that reflect real physics research Creative problem-solving skills across physics domains Familiarity with modeling and approximation techniques in physics Strong written English skills (C1 level or higher) Workflow Apply Complete qualification steps Join a project Carry out assigned tasks Receive payment Time Commitment During active project phases, expect to spend approximately 10–20 hours per week. Actual workload may vary depending on project requirements and timing. Compensation Earn up to $45 per hour, depending on expertise and pace. Rates may differ for other projects on the platform according to their complexity and requirements. Location This is a remote role based in Stuttgart, Baden-Württemberg, Germany.

This freelance, remote role is open to candidates based in Stuttgart, Baden-Württemberg, Germany. Toloka connects skilled individuals with online projects that support artificial intelligence development. Annotators and trainers contribute to generative AI projects for leading technology companies, helping improve the quality and accuracy of AI systems. Role overview Freelance AI Trainers and Annotators (English) play a key part in evaluating and labeling data for AI training. Assignments may include reviewing AI-generated content, checking for factual accuracy, or comparing different responses. The types of tasks available will change according to current project needs and your profile. What you will do Review and assess data, including text, images, and videos Label or categorize content following specific project guidelines Identify and report materials that are factually incorrect, sensitive, inappropriate, or unclear Project details This is a project-based freelance position. Tasks are assigned when projects are active. Participation depends on your profile and current opportunities available through Toloka. Compensation Compensation varies by project. AI trainers can earn up to $22 per hour. How to apply Submit your resume in English and include your English proficiency level.

Role Overview Toloka AI is looking for a freelance Mathematics and Python Specialist to support AI education initiatives. This remote contract role is based in Stuttgart, Baden-Württemberg, Germany. What You Will Do Deliver training sessions focused on Mathematics and Python for AI applications Create interactive learning materials to help students grasp key concepts Support learners as they develop their skills in these subjects Who We’re Looking For Strong background in Mathematics and Python Experience teaching or training others, especially in AI-related topics Ability to design engaging, accessible educational content

Please submit your CV in English and indicate your English proficiency level. About Mindrift Mindrift connects specialists with project-based AI work for major tech companies. The focus is on testing, evaluating, and improving AI systems. This is a project-based role, not a permanent position. Role Overview As a Freelance AI Trainer (Research Physicist), contributors support AI development by creating and documenting advanced physics problems. Each project brings new challenges and may involve: Designing original optics problems that mirror real-world research workflows Ensuring problems require computational approaches, not just manual calculations Developing challenges across mechanics, electromagnetism, thermodynamics, and quantum mechanics Basing problems on actual research questions or practical scenarios in optics and physics Writing clear problem statements and providing verified solutions Qualifications Degree in Physics (Theoretical, Experimental, or Computational) or a related field At least 2 years of experience in applied physics, research, or teaching Familiarity with numerical simulation techniques Ability to design problems that reflect real research processes Creative problem-solving skills across physics disciplines Knowledge of physics modeling and approximation methods Strong written English skills (C1 level or above) How to Apply Apply Pass the qualification process Join a project Complete assigned tasks Receive compensation Time Commitment Expect to dedicate about 10–20 hours per week during active project periods. Actual hours may vary depending on project needs. This estimate does not guarantee a specific workload. Compensation Earnings can reach up to $45 per hour, depending on expertise and contribution speed. Compensation varies by project based on scope, complexity, and required skills. Other projects on the Mindrift platform may offer different rates depending on their requirements. Location: Remote , Stuttgart, Baden-Württemberg, Germany

About the Tendem Project at Mindrift Mindrift connects specialists with AI projects from technology leaders worldwide. For the Tendem project (https://tendem.ai/), the team is building intelligent automation systems that blend Large Language Models (LLMs) with human expertise. The goal: scalable, dependable workflows that enhance content generation, communication, and automation. Role Overview: Freelance AI Workflow Engineer This part-time, remote contract role focuses on designing and refining LLM-powered workflows for the Tendem project. Internally called "AI Pilot," this position partners with Tendem Agents to manage repetitive tasks and develop automation solutions. The work centers on prompt engineering, workflow design, and quality control to ensure reliable outputs. Key Responsibilities Design and evaluate workflows that use LLM APIs (such as OpenAI, Claude, Gemini, or similar). Build and improve AI automation systems for content generation, enrichment, and communication. Develop and refine prompt frameworks to deliver consistent, high-quality results. Create systems for personalized automated outreach, including email, messaging flows, and lead enrichment. Structure and preprocess data to support stable model performance and trustworthy outputs. Analyze and optimize AI-driven pipelines for scalability, efficiency, and accuracy. Who This Role Suits This contract is a fit for those with hands-on experience in LLM integrations, prompt design, and automating AI systems. Structured thinking and attention to quality are essential. The position is remote, based in Stuttgart, Baden-Württemberg, Germany, but open to candidates working from anywhere.

Toloka AI seeks a Freelance Machine Learning Developer to work remotely from Stuttgart, Baden-Württemberg, Germany. This position centers on building and deploying machine learning models that directly support Toloka AI's products. Key responsibilities Create machine learning models tailored to real product requirements Implement technical solutions that drive ongoing improvements Work closely with colleagues to strengthen the company's AI capabilities Location This is a remote role based in Stuttgart, Baden-Württemberg, Germany.

Please submit your CV in English and specify your English proficiency level. Mindrift, in partnership with toloka-ai, offers project-based freelance work for experienced data scientists. This role centers on testing, evaluating, and enhancing AI systems for leading technology companies. All assignments are project-based, not permanent employment. Key Responsibilities Design original data science challenges that reflect analytical work from sectors like telecom, finance, government, e-commerce, and healthcare. Create tasks requiring advanced Python programming, using libraries such as Pandas, Numpy, Scipy, scikit-learn, Statsmodels, Matplotlib, and Seaborn. Develop problems that require significant computational effort, making manual solutions impractical. Formulate scenarios involving complex data processing, statistical analysis, feature engineering, predictive modeling, and extracting business insights. Ensure each challenge is deterministic and reproducible, using fixed random seeds when needed. Base tasks on real business needs, including customer analytics, risk assessment, fraud detection, forecasting, optimization, and operational efficiency. Build end-to-end challenges covering the full data science workflow: data ingestion, cleaning, exploratory analysis, modeling, validation, and deployment considerations. Include scenarios that require scalable computational methods for large datasets. Validate solutions in Python using standard data science libraries and statistical techniques. Write clear documentation for each problem, providing business context and verified solutions. Requirements Minimum 5 years of hands-on data science experience with measurable business outcomes. Portfolio of completed projects or publications showing real-world problem-solving. Advanced Python programming skills for data science (pandas, numpy, scipy, scikit-learn, statsmodels). Expertise in statistical analysis and machine learning, with strong knowledge of algorithms and methodologies. Proficient SQL skills for data analysis and manipulation. Familiarity with GenAI technologies, including LLMs, RAG, prompt engineering, and vector databases. Understanding of MLOps and model deployment workflows. Knowledge of frameworks like TensorFlow, PyTorch, and LangChain. Strong written English skills at C1 level or higher. How to Get Started Apply Complete qualification(s) Join a project Finish assigned tasks Receive compensation Location: Remote , Stuttgart, Baden-Württemberg, Germany

Please submit your CV in English and indicate your English proficiency level. About Toloka AI & Mindrift Mindrift connects skilled professionals with project-based roles at top technology companies, focusing on testing, evaluating, and improving AI systems. This Senior Python Systems Developer position is project-based, not a permanent hire. Role Overview This remote role calls for a Senior Python Developer with deep experience in functional testing. The work centers on building black box tests for large codebases, managing Docker environments, and using language models to interpret code in languages such as C, Rust, and Go. The position also involves translating migration requirements and leveraging tools like Roo Code and Claude Code to streamline development. What You Will Do Design and implement functional black box tests for codebases written in various languages. Set up and manage Docker environments to ensure fully reproducible builds and testing across platforms. Monitor code coverage and establish automated scoring to meet industry benchmarks. Use large language models (LLMs) such as Roo Code and Claude to accelerate development, automate repetitive work, and improve code quality. Requirements At least 5 years as a Software Engineer, with a focus on Python. Strong experience with pytest, including fixtures, session-scoped testing, and timeouts, as well as designing black box functional tests for CLI tools. Advanced knowledge of Docker: writing reproducible Dockerfiles, handling user contexts, and creating secure workspaces. Expertise in Linux and Bash scripting, plus debugging inside containers. Comfort with modern Python tooling (uv, pyproject.toml, packaging). Ability to read and understand code in C, C++, Rust, or Go using LLMs. Hands-on experience using LLMs (Claude Code, Roo Code, Cursor) to speed up development and generate tests. English proficiency at B2 level or above. Preferred Qualifications Experience with agent evaluation platforms and MCP CLI. Key Tools & Technologies Python (pytest, uv, Pillow), Docker, Bash, Git Submodules, C/C++/Rust/Go (for code reading), Dagger, GitHub Codespaces, LLMs (Claude Code, Roo Code, Cursor), coverage.py, gcov, kcov. Benefits & Work Arrangement Freelance, project-based engagement via Mindrift (powered by Toloka AI). Fully remote and flexible schedule. Choose your hours and workload (20-30 hours per week). Compensation varies by project scope and expertise, with potential earnings up to $50 per hour for this project.

Please submit your CV in English and include your English proficiency level. About Mindrift Mindrift connects specialists with project-based AI work at leading technology companies. Projects focus on improving, evaluating, and testing AI systems. This is a project-based, non-permanent role. Role Overview As a Freelance AI Trainer (Research Physicist with Python Skills), each project brings new challenges. Tasks may include: Designing computational physics problems that mirror real research workflows Creating challenges that require Python programming to solve, using libraries such as Numpy, SciPy, and Sympy Ensuring problems are computationally intensive and not easily solved by hand Developing problems that demand complex reasoning in mechanics, electromagnetism, thermodynamics, and quantum mechanics Basing challenges on real research questions or practical applications in physics Validating solutions in Python with established physics simulation libraries Documenting problem statements clearly and providing verified answers Desired Qualifications Degree in Physics (Theoretical, Experimental, or Computational) or a related field Proficiency in Python for numerical validation; experience with MATLAB, R, C, SQL, Numpy, Pandas, SciPy, or similar libraries is also considered At least 2 years of professional experience (applied, research, or teaching) Familiarity with numerical simulation methods Ability to design problems reflecting authentic physics research Creative problem-solving skills across physics domains Understanding of physics modeling and approximation techniques Excellent written English (C1+ level) How Projects Work Submit your application Pass qualification steps Join a project Complete assigned tasks Receive payment Project Commitment Estimated workload is 10–20 hours per week during active project phases. Actual hours may vary depending on project needs, and no specific workload is guaranteed. Compensation Contributors may earn up to $45 per hour, depending on expertise and pace. Rates can vary by project based on scope, complexity, and required qualifications. Other projects on the platform may offer different compensation levels. Location This role is remote and open to candidates based in Stuttgart, Baden-Württemberg, Germany.

Please submit your CV in English and highlight your proficiency in the language. About Mindrift Mindrift connects skilled specialists with project-based AI work for leading technology companies. Projects focus on testing, evaluating, and improving AI systems. Participation is on a freelance, project-by-project basis, not permanent employment. Role Overview: Quantum Research Scientist with Python – Freelance AI Trainer This freelance role involves contributing to unique projects that challenge and expand AI capabilities in computational physics. Each assignment brings new tasks, including: Designing advanced computational physics problems that mirror real research workflows Creating challenges requiring Python programming, using tools such as Numpy, SciPy, and SymPy Ensuring problems are computationally intensive and not solvable manually within days or weeks Developing problems involving complex reasoning in mechanics, electromagnetism, thermodynamics, and quantum mechanics Drawing inspiration from real research challenges or practical physics applications Verifying solutions with Python and standard physics simulation libraries Documenting problem statements clearly and providing verified solutions Who Should Apply This project suits quantum researchers with strong Python skills seeking part-time, non-permanent work. The ideal background includes: Degree in Physics (theoretical, experimental, or computational) or a related field Proficiency in Python for numerical validation (skills in MATLAB, R, C, SQL, Numpy, Pandas, SciPy, or similar domain-specific libraries are also valued) At least 2 years of professional experience in applied research, teaching, or related areas Experience with numerical simulation methods Ability to design problems that reflect authentic physics research workflows Creative problem-solving across various physics domains Familiarity with physics modeling and approximation techniques Strong written English communication skills (C1+ level) How Projects Work The process follows these steps: Apply Pass qualifications Join a project Complete assigned tasks Receive compensation Time Commitment Projects typically require about 10–20 hours per week during active phases, depending on current needs. This is an estimate, not a guaranteed workload. Compensation Contributors can earn up to $45 per hour, depending on expertise and contribution speed. Actual rates vary by project scope, complexity, and required skills. Other projects on the platform may offer different compensation levels. Location This is a remote freelance role based in Stuttgart, Baden-Württemberg, Germany.