Freelance Forward Deployed AI Engineer

Toloka; United States; Remote; USD 150,000 - 230,000; Full time

Job summary

Toloka AI is seeking a Forward Deployed AI Engineer to partner with CTOs, VPs of Engineering, and applied AI teams, translating ambitious model goals into practical, scalable data solutions. You will design data-generation and labeling pipelines, evaluate data quality, and mentor engineers across engagements in a remote-first, globally distributed setup.

You will lead data strategy, ensure robust delivery, and drive improvements in data workflows that power next-generation AI models across

Qualifications

• Executive communication with CTOs and senior leaders.
• Solid understanding of modern LLM development and evaluation.

Responsibilities

• Act as the primary technical counterpart to CTOs, VPs of Engineering, and research leadership.
• Diagnose client data needs and align data strategy with model goals.
• Design and build data-labeling components and AI-driven solutions, integrating them into end-to-end pipelines.
• Own end-to-end delivery including timelines, quality, and commercial margins.

Skills

Executive communication

LLM development knowledge

Hands-on engineering

Client-facing experience

Job description

Company Intro

At Toloka AI we create data that powers leading GenAI models and innovations. We work with frontier labs, big tech, renowned AI startups, enterprises and non-profit research organizations worldwide. We use a combination of Experts + Crowd + Tech Platform to teach AI models to reason and evaluate their efficacy and safety. We have experts in more than 50 different domains—from doctors and lawyers to physicists and engineers—and boast one of the most diverse global crowds, representing over 100 countries and speaking 40+ languages. We are a well-funded startup with an enviable portfolio of clients including Anthropic, Amazon, Microsoft, Poolside, Recraft, and Shopify.

Recently, we secured strategic investment led by Bezos Expeditions and Nebius Group with participation from Mikhail Parakhin, CTO of Shopify and board advisor to leading GenAI companies, who now serves as our Chairman of the Board. Our remote-first team is globally distributed around the world: USA, UK, the Netherlands, Serbia, and more.

About position

The Forward Deployed AI Engineer Engineer is Toloka's most senior client-facing technical role. You will partner directly with CTOs, VPs of Engineering and ML, research leaders, and applied AI teams, helping them solve complex AI data challenges and translate ambitious model goals into practical, scalable solutions.

This is a hands-on role. You will design and build the data-generation, labeling, and evaluation pipelines that power the next generation of AI models. Rather than training the models yourself, you'll help clients identify the right data strategy and own the solution from problem discovery through implementation and delivery.

Beyond delivering client solutions, you'll help shape how Toloka approaches AI solution engineering - establishing best practices, raising the technical bar across engagements, and mentoring Solution Engineers and Technical Consultants.

Why this role is different

Every engagement is different. One week you might be designing evaluation pipelines for frontier foundation models, the next helping an enterprise build domain-specific reasoning datasets or synthetic data workflows.

This role sits at the intersection of consulting, engineering, and AI. You'll work directly with some of the world's leading AI companies, helping shape how next-generation models are trained and evaluated.

What you'll do

Executive partnership

• Act as the primary technical counterpart to CTOs, VPs of Engineering, and research/engineering leadership.
• Lead executive conversations using a structured, answer-first (BLUF) approach.
• Manage escalations and expectations with composure and integrity.
• Build long-term trusted relationships by recommending evidence-based solutions.

Data strategy & needs diagnosis

• Ask excellent questions to uncover the client's real need - the "question behind the question". Understand their model, which metrics they want to move, and how they intend to train or evaluate it.
• Draw on a solid understanding of how LLMs are trained and fine-tuned to have credible conversations with their technical leaders, understand their data strategy, and proactively propose the data that will solve their problem - with options and rationale ("based on your goal, you likely need this, or this").

Solution engineering & delivery (hands-on)

• Design and build the data solutions yourself: configure data-labeling components and quality controls, develop user interfaces and AI-driven solutions (e.g. agentic systems, RAG, synthetic data generation), and integrate them into automated, multi-stage pipelines that produce data for AI training and evaluation.
• Architect and reason about complex, multi-stage solutions end to end; run experiments to prove the pipeline delivers data of the required quality and speed; iterate from MVP toward production.
• Provide technical leadership across multiple client engagements, establish reusable engineering standards and best practices, and mentor Solution Engineers and Technical Consultants to raise the overall technical bar of the organization.

Commercial ownership

• Own delivery end to end - timelines, quality, and scalability - and the commercial outcome : Gross Margin and Contribution Margin, with the levers, trade-offs, and next checkpoints named, not just described.
• Identify and drive expansion opportunities.

About You

No one is expected to meet every requirement below. If you bring a strong combination of client-facing communication, AI/LLM expertise, and hands-on engineering experience, we'd love to hear from you.

• Executive communication. You can confidently lead conversations with CTOs, VPs, and senior technical stakeholders - clear, concise, structured, persuasive, and calm under pressure. Professional English (C1+) is essential.
• Strong understanding of modern LLM development. You understand how modern LLMs are trained, fine-tuned, and evaluated (including SFT, RLHF/RLAIF, …