Lantide Data

Use Cases

For analysis that must be explained, rerun, or handed off after the first answer.

Recurring Analysis and Reporting

New data, same definitions, rerunnable workflow

The work to be done

Operations, finance, and product teams receive a new CSV, Excel file, or database snapshot every week or month and need to update the same analysis with consistent definitions.

Best for

Analysts and operations teams that regularly refresh metrics, anomaly analysis, or management reports.

Why the current approach breaks

Chat-based analysis starts over in every session. Even when the question is unchanged, users must restate fields, definitions, filters, joins, and validation, with no clear proof that the new result follows the last formal version.

How Lantide turns it into an analysis asset

  1. 1

    Carry forward the working environment

    A Workspace bundle can carry the Plan, SQL evidence, Report, Agent query ledger, MCP cache, and selected local data into the next cycle.

  2. 2

    Load the latest data

    Reconnect a new file or refresh a source, then verify schema and reporting period before execution.

  3. 3

    Rerun by lineage and produce a new Report

    Use Source Run and dependencies to refresh affected steps while preserving evidence for the new result.

Structured Analysis Projects

A/B tests, funnels, anomaly detection

The work to be done

E-commerce, SaaS, gaming, and subscription product teams rely heavily on data decisions and frequently analyze A/B tests, conversion funnels, retention changes, and anomaly metrics.

Best for

PMs, analysts, and business teams evaluating product, growth, or operations decisions.

Why the current approach breaks

A product team wants to know whether a new onboarding flow improves activation, but chat-based AI scatters conclusions across conversations while SQL, assumptions, filters, and analysis logic are hard to trace. When leadership asks where a conclusion came from, the analyst must dig through chat history and SQL tabs.

How Lantide turns it into an analysis asset

  1. 1

    Analyst and Agent co-author the Plan

    The analyst and Agent first turn the objective, data scope, success metrics, and methodology into a reviewable Plan so PMs, leaders, and analysts align before execution.

  2. 2

    Review with annotations

    Reviewers can comment directly on the Plan, such as excluding internal test accounts or splitting new and returning users, with every revision preserved in document context.

  3. 3

    Execute to produce Report

    Only after the Plan is confirmed does the team execute the analysis; Lantide locks the Executed Plan and produces a Report so conclusions, methods, and versions remain traceable.

Privacy & Compliance

Local-first, sensitive data

The work to be done

Healthcare, finance, and multinational retail teams operating under regulatory, contractual, or information-security requirements.

Best for

Analytics, security, and risk teams that need explicit data and model boundaries.

Why the current approach breaks

The company already has a secure Azure endpoint and wants to adopt AI to improve analysis efficiency, but many cloud AI assistants still require sensitive customer data or transaction CSVs to be uploaded into an external environment. Data egress, control boundaries, and auditability remain major security and compliance concerns.

How Lantide turns it into an analysis asset

  1. 1

    Load data in local DuckDB

    Analysts designate a local folder as the Workspace, and DuckDB reads, queries, and caches the data on the user's device.

  2. 2

    Disable or approve knowledge injection

    Any new business rules summarized by the model from conversations stay locked in the Queued Knowledge approval queue and are never injected into AI context without review.

  3. 3

    Results stay in workspace

    With a SELECT-only read guard, AI can only read data and create temporary computation caches. It cannot write back to original data, reducing tampering, accidental deletion, and unauthorized modification risks.

Multi-step SQL Pipelines

Cache chaining, lineage tracing

The work to be done

E-commerce operations, supply chain, financial analysis, and data teams need to connect raw transactions, user events, order details, and product dimensions into multi-step analysis workflows.

Best for

Data analysts maintaining multi-step SQL, recurring updates, and intermediate results.

Why the current approach breaks

This work is rarely a single SQL query. In the AI era, SQL is generated faster, longer, and in greater volume, but cache freshness, query dependencies, rerun order, and lineage become harder to manage and trace.

How Lantide turns it into an analysis asset

  1. 1

    Build multi-tab SQL with cache

    Analysts can split each query step into clear SQL tabs and cache intermediate results as reusable tables instead of recalculating from scratch every time.

  2. 2

    Source Run recalculates via DAG

    When upstream data or a query changes, Source Run uses the DAG to determine dependency order, especially for recurring updates that continue to branch into deeper exploratory queries.

  3. 3

    View lineage and source SQL

    Every result can be traced back to source SQL and upstream cache, helping analysts answer where a number came from and deliver with confidence.

Cross-source Analytics

Files + external DB + MCP

The work to be done

Retail, manufacturing, SaaS, finance, and operations teams often have data scattered across local Excel / CSV files, internal PostgreSQL / MySQL databases, third-party exports, and MCP tools.

Best for

Operations and data teams combining files, databases, and external services.

Why the current approach breaks

A team wants to analyze whether ad campaigns are causing stockouts in specific regions, but the data lives across ad platform CSV exports, ERP databases, support systems, and local Excel files. Moving data, converting formats, and aligning fields consume time before analysis begins.

How Lantide turns it into an analysis asset

  1. 1

    Load local files

    Analysts can load CSV, Excel, Parquet, and other local files directly into the workspace and quickly turn them into queryable tables.

  2. 2

    ATTACH external databases

    Operational data in PostgreSQL or MySQL can be connected through DuckDB ATTACH without manual export and re-import.

  3. 3

    Query with shared logical table names

    AI agents and analysts use the same logical table names and unified query layer, making cross-source JOINs natural while reducing data movement and schema confusion.

Stakeholder Deliverables

Reports for bosses/clients

The work to be done

Consulting firms, product teams, operations teams, and internal analytics units often need to deliver analysis results to executives, clients, investors, or cross-functional stakeholders.

Best for

Consultants and analytics teams delivering work to leaders, clients, or cross-functional stakeholders.

Why the current approach breaks

Analysts can get results using SQL and notebooks, but leaders and clients do not read query processes or chat history. They need reports with context, conclusions, charts, and recommendations; manually copying analysis into slides or documents is slow and error-prone.

How Lantide turns it into an analysis asset

  1. 1

    Execute to produce Report

    After analysis is complete, Lantide produces a structured Report from the reviewed Plan, preserving methods, findings, evidence, and recommendations.

  2. 2

    Generate HTML report

    Analysts can turn the Markdown Report into a standalone HTML report and use Quick Edit in the browser to tune the tone and layout for leaders or clients.

  3. 3

    Link the External Delivery

    Record the Report's P4, KM, URL, or Other destination as pointer metadata. Lantide does not upload, verify, synchronize, or delete the external content.

Cross-Workspace Verification

Snapshot-safe evidence, one writer Workspace

The work to be done

An external Agent manages work in one Workspace but needs to verify an approved Plan, Report, reference, delivery pointer, or evidence lineage from another.

Best for

Teams and Agent operators reviewing related work across separate Workspaces.

Why the current approach breaks

Switching environments interrupts active work, while broad cross-Workspace access could expose live sources or create competing writers.

How Lantide turns it into an analysis asset

  1. 1

    Keep one writer Workspace

    The Agent remains bound to its selected Workspace for deep work and every write.

  2. 2

    Read snapshot-safe artifacts on demand

    A per-call read can inspect supported Plans, Reports, references, HTML artifacts, External Delivery lists, and evidence lineage in another Workspace.

  3. 3

    Preserve the boundary

    The cross-Workspace path does not switch the GUI or writer, and it does not permit live queries, source access, drafts, pending operations, or writes.

External Agent Handoffs

Portable context, governed Agent work

The work to be done

A team hands recurring or long-running analysis to another person, machine, or MCP Agent without flattening the work into chat transcripts.

Best for

Analysts and platform teams coordinating built-in and external Agents.

Why the current approach breaks

Long investigations exceed short conversation windows, and a new Agent can miss prior query steps, source assumptions, or the location of a delivered Report.

How Lantide turns it into an analysis asset

  1. 1

    Preserve Plan-local execution

    Full execution snapshots keep formal query and Source Run steps even beyond the 50-step conversation context window.

  2. 2

    Transfer the Workspace when needed

    Use a Workspace bundle to move local working data, MCP cache, and the Agent query ledger under explicit export and import controls.

  3. 3

    Resume with governed access

    The next Agent can inspect formal artifacts, use External Delivery pointers, and continue under the selected Workspace's read, Execute, and approval boundaries.