DX vs Toloka

DX AI is a conversational interface for exploring an organization's engineering data held in DX, described by the vendor as 'the copilot for engineering leaders.' It answers natural-language questions with generated charts, summarizes qualitative snapshot data into themes and sentiment, and diagnoses the drivers behind metric changes. Its data can also be streamed into external LLM clients via the DX MCP server.

The DX MCP server exposes a customer's DX data to MCP-compatible AI clients such as Claude, Cursor and Devin Desktop. It is deliberately read-only, offering tools to list and retrieve software catalog entities, scorecards and initiatives, and teams, plus execution of SQL queries against the DX Data Cloud PostgreSQL database. It is available as a DX-hosted remote endpoint at ai.getdx.com/mcp or run locally from DX's open-source dx-mcp-server repository.

AI Code Insights measures what AI coding agents produce inside an engineering organization, tracking AI-generated code by commit, PR, team, agent and repo from IDE to production. It attributes authorship by monitoring supported coding agents' edits and links them to commits, pull requests and deployments, reporting adoption, delivery velocity, code-quality signals and session-level agent performance. It includes the Agent Experience report, which scores requirements clarity, steering and task scope from the agent's perspective.

Command-line interface to a DX instance, built so coding agents can drive it: it ships an agent skill telling an agent when and how to call the CLI, and covers the same catalog, scorecard and self-service operations a human uses the dashboard for.

Fabric is DX's context layer for AI agents, marketed as 'the context engine for AI-native engineering.' It turns an organization's software catalog into structured, live context that agents can query, and adds scorecards and self-service automation so agents can create services, provision infrastructure and resolve failing standards checks. Context is exposed to any MCP-compatible agent via the DX MCP server.

A hybrid AI-plus-human agent that takes a delegated task - research, data analysis, copywriting, design or development - has AI do the first pass, then routes it to one of 10,000+ vetted experts for verification and multi-layer QA, returning results in 2-24 hours.

An independent evaluation platform that ranks frontier LLMs on agentic tool-use tasks using private, non-contaminated benchmarks across industry domains, scored on a pass^5 reliability metric, with the underlying RL Gyms and evaluation datasets available to license.

A self-serve service that lowers per-request inference cost through two tools - fine-tuning LoRA adapters on frozen Qwen3 base models to replace a frontier API on a narrow task, and prompt gisting that compresses long instruction prefixes into learned tokens.

Toloka's catalogue of ready-made training datasets sold outright — three named at the time of writing (Tau-bench Dataset Extension, University-level Math Reasoning, Multimodal Conversations) — as distinct from the custom data work its Platform sells.

Human-in-the-loop data programs -- demonstrations, annotation and evaluation -- for training robotics and physical AI systems.

A self-serve platform where an AI agent turns a described data goal into a full human-annotation pipeline - RLHF and preference data, data collection, instruction tuning, model evaluation, synthetic-data validation and content-moderation QA - with LLM-based quality checks on the output.