Altilia vs Dataiku (2026)

Pre-built connectors and an API-first architecture let AI assistants and robots be embedded across the customer's tech stack, from legacy systems to modern cloud applications, and agents take action on data. No individual connector, integration catalog or connector count is named for any application class. The platform is distributed through the Microsoft commercial marketplace. Documents and data are ingested from any source or format, including PDFs, emails, handwritten notes, scanned images and legacy databases.

Sourcealtilia.ai home, platform, why-altilia and modules pagesread 2026-08-31

Agents call managed tools, API endpoints, OpenAPI specs, local and remote MCP servers, Knowledge Bank search, LLM Mesh queries, web search, inline code and plugin tools, over connections secured per group.

Sourcedoc.dataiku.com/dss/latest/agents/index.htmlread 2026-09-27

Agent workflows are designed with visual scripting, automating and sequencing actions and connecting data sources with fine-tuned models, chains and processing functions, alongside skills, which are process orchestration tasks built in the low-code interface.

The platform provides governable agentic orchestration and ontology-driven workflows that coordinate agents across departments, with agents shipping as assistants or autonomous robots within the same workflow model. Independent analysts describe the agents-and-skills combination as providing robotic process automation capabilities. No branching, looping, conditional or parallel control-flow construct is named, and no failure or retry behavior is described.

Sourcealtilia.ai platform, agents and home pages, Intellyx analyst briefread 2026-10-05

Structured visual agents define sequences with routing, delegation to other agents, parallel and for each branches, agentic loops and reflection blocks, alongside code agents and external agents from Snowflake Cortex, Databricks, AWS Bedrock and Google Vertex AI; Agent Hub can orchestrate a query across several agents.

Sourcedoc.dataiku.com/dss/latest/agents/structured-visual-agents/index.htmlread 2026-09-27

Agents are triggered by document and data ingestion, with the platform ingesting and structuring data from any source or format, including legacy databases, emails, scanned images and forms, and an API-first architecture lets agents embedded in customer applications be invoked programmatically. No scheduled, cron or recurrence capability, inbound webhook or external event subscription is named, and no chat, email, messaging or embedded conversational channel reaches an agent beyond the in-platform Insights assistant.

Sourcealtilia.ai platform, home and modules pagesread 2026-08-31

Agents run inside automation scenarios, which start on time based, dataset change, SQL query change, scenario completion or Python triggers without a person, and users reach agents in Slack (DM or mention), Microsoft Teams, Agent Hub, Agent Chat and through APIs.

Sourcedoc.dataiku.com/dss/latest/scenarios/index.htmlread 2026-09-27

The Knowledge Graph is the platform's core component, aggregating and organizing all company information and unifying structured, unstructured and physical document sources into a single coherent database. A hybrid neuro-symbolic architecture combines large and small language models with the graph and is described as GraphRAG with traversal reasoning.

Ingestion covers any source or format, including long complex documents, standardized forms, scanned images, handwritten notes, emails and legacy databases, transformed into semantic objects. Users interrogate the knowledge base in natural language and receive fact-based answers with source attribution, and RAG techniques retrieve and fine-tune LLMs and SLMs against enterprise knowledge. Accuracy above ninety-five percent on domain-specific tasks is vendor-reported.

Sourcealtilia.ai platform, home and why-altilia pagesread 2026-08-31

Knowledge Banks embed datasets and documents into vector stores that agents search through a Knowledge Bank Search tool with static, dynamic and agent inferred filters and document level security, with GraphRAG and automated RAG optimization also documented.

Sourcedoc.dataiku.com/dss/latest/agents/tools/knowledge-bank-search.htmlread 2026-09-27

There is no memory store, session state or cross-run context. Continuous learning from human feedback, which Altilia describes as building training datasets, feeds the training loop.

Sourcealtilia.ai/platformread 2026-09-29

Long term memory blocks store facts, such as preferences, and episodes from past conversations in memory banks, kept separately per user, shared by agents in a project through a default bank or split into named banks, and retrieved by LLM filtering and semantic search in later conversations. No retention period or delete path is stated.

Sourcedoc.dataiku.com/dss/latest/agents/structured-visual-agents/blocks/long-term-memory.htmlread 2026-09-27

In a review and validation stage, people review and validate data classifications, extractions and AI-generated answers, and the feedback of human experts is integrated to enhance accuracy and control at key stages, with reviewed corrections feeding continuous learning. Agents ship in two forms, assistants that support human decision-making and robots that automate processes autonomously, so the autonomy level is a design-time choice. No confidence threshold routing uncertain items to review, action allowlist, or scope or spend limit is described.

Sourcealtilia.ai platform, agents and home pagesread 2026-10-05

Human approval set on a tool makes a visual agent pause before calling it and ask the user to confirm, optionally letting the person reviewing edit the tool inputs (not inside For Each, Parallel or Reflection sub sequences), and guardrails cover PII, prompt injection, toxicity, topic boundaries and bias.

Sourcedoc.dataiku.com/dss/latest/agents/tools/human-approval.htmlread 2026-09-27

Identity and Access provides secure access with RBAC, SSO and multi-tenancy, with guardrails on model and agent behavior, and the site footer shows ISO/IEC 27001, ISO/IEC 27017 and ISO/IEC 27018, plus ISO 9001, as certified. The security page describes the management system as aligned with those standards, which is consistent with holding the certificates, but no certificate or registrar is named.

Sourcealtilia.ai/platformread 2026-09-29

Dataiku offers SSO through SAML, OIDC, LDAP and Kerberos, MFA, user profiles, per project group and user permissions, connection security, user secrets and an audit trail. Its trust page lists ISO 27001:2022, ISO 27701, ISO 42001 and ISO 9001 certification, SOC 1 and SOC 2 Type II for Dataiku Cloud, and a HIPAA report.

Sourcedataiku.com/legal/trustread 2026-09-27

Auditability gives an end-to-end audit trail of every action and decision, and Observability gives real-time logging, monitoring and alerting, built into every platform module, alongside workflow and agent orchestration run step by step with monitoring and human validation.

Sourcealtilia.ai/platformread 2026-09-29

Every LLM Mesh call returns a nested trace of the agent's LLM calls, guardrail steps, inputs, outputs, tokens and cost, viewed in the Trace Explorer (tree, timeline and explorer views), and the evaluation trajectory explorer replays tool calls in order; an audit trail records user actions.

Sourcedoc.dataiku.com/dss/latest/agents/tracing.htmlread 2026-09-27

SaaS, private cloud and on-premises deployment are offered, which Altilia calls deployment sovereignty, with domain-tuned small and large language models served locally so inference runs inside the deployment rather than being sent to a hosted provider.

Data is kept within the customer's secure infrastructure, and customers retain ownership of data, models and assets. The company positions itself as a sovereign European platform competing with US hyperscalers.

No region selection within the SaaS offering, installation requirement, or statement of which components run customer-side in the on-premises mode is described.

Sourcealtilia.ai why-altilia, home and platform pagesread 2026-08-31

Dataiku runs as fully managed Dataiku Cloud or is installed on the customer's own infrastructure (Linux, macOS, experimental Windows) with hybrid options through sales, with installation, cloud and container deployment guides; locally running Hugging Face models keep inference in house.

Sourcedataiku.com/product/get-startedread 2026-09-27

Customers can deploy ready-to-use agents for common use cases, including extracting data from invoices, orders and transport documents, or use the integrated development environment to create bespoke agents. A broader specialized set covers document classification by semantic context, data extraction from unstructured sources, semantic search and question answering with source attribution, summarization into reports, and generation of emails and operational content, and blueprints for common use cases are referenced.

Altilia lists 45 named agents across eight industries and a cross industry group, among them Credit Memo Drafting, Customer Onboarding Data Capture, Secured NPL Data Acquisition, Insurance Claims Management, the DORA Compliance Agent, Hospital Order Management and the Enterprise Search Agent, each with a line on the job it does. Customers reach them through a demo request rather than a self serve deploy button.

Sourcealtilia.ai/agents, platform and modules pagesread 2026-10-05

Dataiku Solutions include named GenAI and agent solutions for separate jobs, Dynamic Selling Assistant, Supplier Management Assistant, Clinical Trial Intelligence Assistant, Medical Entity Extraction Assistant and Agentic Insights; Agent Hub distributes an organization's own governed agents and prompt library.

Sourceknowledge.dataiku.com/latest/solutions/index.htmlread 2026-09-27

The platform is model agnostic and runs any commercial or open-source model, with domain-tuned small and large language models served locally and model selection driven by cost and data type, which Altilia frames as customers owning their models.

RAG techniques retrieve and fine-tune LLMs and SLMs against enterprise knowledge, and AI agents trained on customer data remain the customer's intellectual property. Altilia positions model portability as the basis of its sovereign European alternative to hyperscaler platforms. No named provider list, model picker interface or automatic runtime routing between models is described.

Sourcealtilia.ai platform, why-altilia and home pagesread 2026-08-31

Administrators define LLM Mesh connections to OpenAI, Azure OpenAI, Anthropic, AWS Bedrock and SageMaker, Microsoft Foundry, Google Vertex, Mistral, Cohere, NVIDIA NIM, Snowflake Cortex, Databricks foundation model APIs and locally running Hugging Face models, and agents pick from them, with cost control and rate limiting documented.

Sourcedoc.dataiku.com/dss/latest/generative-ai/llm-connections.htmlread 2026-09-27

AI agents connect directly into the customer's operations and applications through an API-first architecture, embedded across the tech stack from legacy systems to modern cloud applications. Simple SDKs are one of three first-class routes for building agents, alongside natural language and visual scripting. No API reference, endpoint list, authentication documentation, SDK package, developer portal or MCP server is published.

Sourcealtilia.ai home, platform and why-altilia pagesread 2026-08-31

Agents are exposed as A2A servers at a per agent public API endpoint with well known URI discovery and bearer API key auth, in JSON-RPC and SSE modes, beside a documented MCP server, the public REST API, the Python API and LangChain integration on developer.dataiku.com, and a plugin system.

Sourcedoc.dataiku.com/dss/latest/agents/a2a.htmlread 2026-09-27

Monitoring and tracking of model performance spot inefficiencies, with continuous learning and improvement over time, alongside human review and validation of data classifications, extractions and AI-generated answers whose corrections feed the training loop. There is no held-out test set, recorded expected outputs, scored run, evaluation harness or comparison between agent versions. Accuracy above ninety-five percent on domain-specific tasks is vendor-reported.

Sourcealtilia.ai platform, why-altilia and home pagesread 2026-08-31

The Evaluate Agent recipe scores answers and the path taken, with ordered and unordered tool call checks, RAGAS based LLM judges, BERTScore and custom metrics, stores results for side by side version comparison, and Agent Review lets builders and SMEs run test cases against an agent version with human verdicts overriding the judge.

Sourcedoc.dataiku.com/dss/latest/agents/evaluation.htmlread 2026-09-27

Altilia calls its autonomous agents robots, as distinct from assistants, and independent analysts describe the agents-and-skills combination as providing robotic process automation capabilities.

The mechanism is agents classifying, extracting, summarizing and generating against documents and a knowledge graph, reaching systems through an API-first architecture and pre-built connectors, not through interface operation. Computer vision and document analysis and recognition read scanned images and handwritten notes as part of document intake. There is no browser control, navigation, form filling, screen interaction, visual grounding or desktop automation.

Sourcealtilia.ai platform, agents and home pages, Intellyx analyst briefread 2026-10-05

The agent tool list covers APIs, MCP, knowledge search, code and an LLM side web search tool, with no browser or computer use tool documented.

Sourcedoc.dataiku.com/dss/latest/agents/index.htmlread 2026-09-27