Pipelines, models,
and AI — on one canvas.
OpticIQ is a drag-and-drop workspace where teams build data workflows, run AI models, and ship results — without writing a line of code.
Built in-house by HEMOdata · Invite-only · No credit card required
The problem teams face today
Teams can ask questions but the chatbot can't see your systems, run your numbers, or take action afterwards. Answers end where the work begins.
Every meaningful pipeline today is built in Python or SQL. The people who know the business can't build for the business.
One tool for ETL. Another for dashboards. Another for AI. None of them share data — so your team spends its day reconciling, not deciding.
From source to outcome.
Four steps. One canvas. The whole flow is visual, runnable, and shared with your team from day one.
Your databases, M365 files, APIs, spreadsheets, and inboxes — all in one palette. Drag and drop, no credentials to hardcode.
Filter, aggregate, join, pivot, dedupe, type-cast. Every operation is a node — each one showing you its row count and timing.
Forecast, detect anomalies, classify, score sentiment — or call Claude for reasoning, extraction, or structured output. All in the same flow.
Live dashboards. Write-backs to your systems of record. Personalised emails and briefings. Webhooks to trigger whatever comes next.
Four kinds of node.
Everything you need.
The palette is organised around intent — what does this step do? — not data type. New users see what's possible without reading the docs.
Connect to databases, files, APIs, and spreadsheets. Filter, group, join, dedupe, and pivot with visual nodes — then write the result wherever it needs to go.
Forecasting, anomaly detection, classification, sentiment scoring, and clustering run entirely inside OpticIQ. No API calls, no token bills, no surprises.
Upload your own PDFs, slides, and notes. Bind them to an AI node. Every answer is rooted in your material and cites the source it came from.
Schedule pipelines. Trigger them from a webhook, a form, or an inbox. Branch on a condition, loop over a list, assert data quality, and alert on failure.
Your one explicit AI node.
Most platforms hide AI calls behind dozens of buttons. OpticIQ does the opposite — AI lives in a single, magenta-bordered node. Cost, behaviour, and output are all visible before you click.
OpticIQ shows you what one run will cost against your dataset before you click. No surprise invoices at the end of the month.
Every Agent returns a defined shape — fields you can filter, write to a database, or pipe into the next step. No prose to parse.
When a row fails, skip it, log it, or fork it to a recovery branch. The whole pipeline doesn't stop for one bad row.
Ground your AI in your own docs.
Upload your policies, playbooks, and reports. Attach them to any Agent. Every answer the AI gives is rooted in your material — not the open internet.
PDFs, Word docs, slides, sheets, and notes. Drag them in through the sidebar.
Attach the library to an Agent node. It searches your docs per row, automatically.
Every answer the Agent gives references the document and chunk it came from.
Choose whether the AI is bound strictly to your docs, or whether it can answer freely when your docs don't cover the question. You pick the tradeoff.
Knowledge search runs entirely inside OpticIQ — no embeddings API, no third-party cost. You only pay for the actual Claude calls in your Agent node.
Connects to your stack.
Pull from the systems your team already uses. Push to the places your team already looks.
Secure and collaborative,
from day one.
The basics that serious teams expect — built in, not planned for a future enterprise tier.
Use what your team already has. Profile photos and identity flow through automatically from M365.
Sign-up is closed by default. People join only when an existing member invites them — no public self-serve.
Owner, Admin, Member. Each role does exactly what its label says — no surprises. Members see only what's been shared with them.
Each run records what was in the canvas, who ran it, what came out, and what failed. Replay it any time.
Cancel a run mid-flight. Resume from where it broke. No more babysitting jobs that should have finished overnight.
The Agent is the only place AI runs. You see the cost before you click — and no other node bills tokens behind your back.
Common questions about OpticIQ.
Quick answers for teams evaluating OpticIQ for the first time.
What is OpticIQ?
OpticIQ is a drag-and-drop workspace where teams build data pipelines, run AI models, and ground Claude in their own documents — without writing code. It puts data sources, transforms, models, and outputs onto a single visual canvas.
Who is OpticIQ for?
Business and operations teams who need to wrangle data into useful outputs but should not have to learn Airflow, dbt, or Python. Anyone who can describe what they want — without needing to learn DAG syntax.
Does OpticIQ require coding?
No. OpticIQ is fully visual. Every pipeline — sources, transforms, models, AI agents, outputs — is built by connecting nodes on a canvas. There is an optional Python Script node for power users, but it is never required.
How does OpticIQ use Claude AI?
OpticIQ has a single, explicit Agent node where Claude is called. The Agent shows cost before you run, returns structured output by a defined schema, and can be grounded in your uploaded documents through a Knowledge Base. No other node in OpticIQ bills tokens — only the Agent.
How is OpticIQ different from ChatGPT?
ChatGPT is a chat interface that answers questions. OpticIQ is a workflow platform that connects to your data systems, runs models, and delivers outcomes — dashboards, write-backs, emails, and webhooks. Where ChatGPT answers, OpticIQ acts.
How is OpticIQ different from Zapier or Make?
Zapier and Make connect apps. OpticIQ is purpose-built for data and AI workflows — with built-in machine learning models, an explicit Claude Agent node, document grounding via Knowledge Bases, and analytics-grade run history with replay and diff.
How is OpticIQ different from dbt or Airflow?
dbt and Airflow are engineer-facing tools that require SQL and Python. OpticIQ is a visual canvas designed for business teams — no YAML, no DAG syntax. It also includes built-in ML models and AI agent support that dbt and Airflow don't provide natively.
Can I upload my own documents?
Yes. Upload PDFs, Word docs, slides, spreadsheets, Markdown, CSV, or text files into a Knowledge Base. Bind that Knowledge Base to a Claude Agent and every answer is rooted in your material, with citations back to the source.
What does OpticIQ connect to?
SQL databases (PostgreSQL, MySQL, SQL Server), Microsoft 365 (OneDrive, SharePoint, mailboxes), spreadsheets (Excel, CSV), REST APIs, and email inboxes. Outputs include live dashboards, database write-backs, emails, webhooks, and files.
Is OpticIQ secure?
Yes. Password or Microsoft 365 single sign-on, invite-only by default, project visibility scoped to your organisation, audit-ready run history for every execution, and AI cost shown up front before any call is made.
How much does OpticIQ cost?
OpticIQ is in soft launch with invite-only early access. Pricing isn't public yet — request early access and we'll be in touch with details.
Who builds OpticIQ?
OpticIQ is built in-house by HEMOdata, a GCC-based data and AI consultancy founded in 2022. The product is 100% owned by HEMOdata with no external IP dependencies.
Ready to build on one canvas?
OpticIQ is in active development. Request early access and we'll be in touch with an invite as slots open.
Built in-house by HEMOdata · No credit card · Invite-only
