olllo

The main problem is that less and less human visit stackoverflow, but not others. So I think it should face to agent but not human in this case.

It should adjust traget users from human to agent. That's to say, to build for agents sharing knowledge.

The "million row limit" in Excel is a relic. Even when using tools like VS Code or Notepad++ for large files, the memory pressure often makes the UI unusable.

DataOlllo is a lightweight viewer designed to stream data directly from disk. It handles massive CSVs and modern formats like Parquet and Arrow natively. It’s been helpful for my workflow when I just need to filter and export a subset of a massive dataset without spinning up a Jupyter notebook.

DataOlllo, Your Private AI Data Analyst — Analyze, Automate, Reuse. DataOlllo helps you explore, analyze, and visualize your data with AI—fully offline and secure. Build reusable workflows, automate complex tasks, and get instant insights. No coding needed, just clicks and natural language.

Try Google Antigravity with Gemini 3.

Cursor and Claude are better than it now.

But I believe the mode of antigravity will continue to grow until it beyonds claude mode.

The agents coperation in antigravity has strong evolution potention

Record every step you take — cleaning, transforming, analyzing, and visualizing data — then save it as a reusable workflow to run anytime.

Hi HN! I’m excited to share the newest version of Data.olllo, a local AI workflow automator for data processing, visualization, and reusable workflows.

What’s new • Next-gen workflow engine: Build, save, and reuse workflows for any dataset • AI-assisted automation: Generate steps or entire pipelines with AI agents • Full local processing: Secure, offline, with no raw data sent to the cloud • Scalable: Works from small CSVs to multi-GB datasets

Why it’s useful • Clean, transform, and visualize data without writing code • Apply saved workflows to similar datasets instantly • Ideal for repetitive tasks, analytics, ecommerce data, logs, or any large dataset

Get it

Website / Download: https://olllo.top

I’d love feedback from the HN community on this updated version, especially on workflow automation and AI-assisted data tasks.

Every week, Shopify sellers download sales CSVs and Facebook ad reports, then spend hours merging them in Excel just to figure out if they actually made money. I got tired of that.

So I built Data.olllo — a local desktop app that lets you open both CSVs, ask an AI assistant things like “calculate profit by campaign per month”, see the code it generates, preview the result, then click “Apply All” to run it safely on your full data. You can even visualize the outcome (income, outcome, profit) as interactive bar charts and save the whole workflow to rerun next week — all locally, no cloud uploads.

It turned what used to be a 2-hour spreadsheet task into a 10-second repeatable workflow.

I’d love feedback from the HN community — on the product, workflow design, or how to make the AI assistant smarter for real-world data tasks.

If you work with large CSV, Excel, or database files, Data.olllo is a new desktop tool that lets you clean, analyze, and visualize datasets locally without uploading to the cloud. - Privacy-first: everything stays on your machine - Lightning fast: handles millions of rows and 100GB+ files - AI assistant: generates pandas, plotly, and p5.js code you can trust and control - No-code interface for all your common data operations

The GitHub repo includes README, setup instructions, and links to guides and demos: https://github.com/olllo-top/data.olllo-helper

Perfect for analysts, researchers, or anyone frustrated with Excel crashing or cloud upload delays.

Data.olllo — a local-first, privacy-focused data analysis tool designed for large datasets. Explore and visualize your data interactively without coding or uploading sensitive information to the cloud. Works offline, fast, and secure.”

I’m the creator of Data.olllo, a desktop data tool designed as a powerful alternative to Excel, built to handle large and complex datasets without the usual crashes or slowdowns.

With Data.olllo, you can open, clean, and visualize CSV, Excel, or JSON files instantly — all offline, with no cloud uploads and zero coding required. Whether you’re working with hundreds or millions of rows, Data.olllo stays fast and responsive.

Key highlights: • Blazing-fast loading and exploration of large datasets • 100% local processing for total data privacy and security • AI-powered natural language commands to clean, merge, and analyze data without formulas or scripts • Multi-core and GPU acceleration to leverage your hardware fully • Intuitive no-code workspace suitable for data analysts, scientists, and anyone frustrated by Excel’s limits

I built Data.olllo to solve the headaches of Excel crashing on big files and the privacy concerns of cloud-based AI tools. If you want a seamless, private, and powerful data analysis experience, check it out.

Try it here: [https://olllo.top/excel-alternative-for-large-datasets]

Happy to answer questions or hear feedback!

Hi HN,

I want to share a cool feature in Data.olllo, a local, offline, no-code data analysis tool with AI chat capabilities.

Handling time and date data can be a pain—timestamps, strings, extracting year/month, or converting formats usually require coding knowledge. With Data.olllo’s AI chat, you just describe what you want in plain English, and the AI instantly generates the correct pandas code inside a process(dfs) function you can run on your dataset.

For example, you can: - Convert Unix timestamps to readable date-times - Extract parts like year and month from date columns - Parse messy string dates into proper datetime objects - Convert datetime back to numeric timestamps - Format datetime columns into any string style you want

Here’s a sample snippet the AI generates for converting a Unix timestamp column:

def process(dfs): df = dfs["df"] df["timestamp"] = pd.to_datetime(df["timestamp"], unit="s") return df

You don’t need to know pandas or write any code yourself — just type your request, and the AI does the heavy lifting, letting you explore and visualize your data faster.

Data.olllo runs 100% locally, so your data stays private, and it can handle millions of rows quickly.

If you often struggle with time data or want a fast way to analyze large CSVs without coding, give Data.olllo a try:

https://olllo.top/convert-format-datetime-ai-chat

Happy to answer questions or get feedback!

— Denis

We all have those moments: a massive dataset, full of potential, but locked behind rows, columns, and hours of manual digging. I built something to change that — not just for myself, but for anyone who’s ever stared at a CSV and thought, “There has to be a better way.”

Data.olllo is a desktop app that lets you talk to your data.

Ask in plain English:

“Which products grew the fastest this year?” “What’s unusual about Q2 performance?” “Show me regional trends for refunds.”

The AI assistant understands your dataset and responds with insights — tables, summaries, even charts. No Python scripts. No cloud latency. No need to upload anything. Your data stays with you. It just becomes smarter.

Behind the scenes, it uses your choice of AI (ChatGPT, Gemini, or even a local LLM), but the goal isn’t just automation — it’s flow. You and your data, in sync.

This isn’t a data tool. It’s a new way of thinking with your information.

Try it here → https://olllo.top/AI-CSV-Analysis

I built a tool called Data.olllo that helps split huge CSV files—like multi-GB datasets—into smaller parts by size, row count, or column value. It’s 100% offline, runs on your desktop, and doesn’t require any coding.

We just published a detailed article explaining how it works: https://olllo.top/articles/article-24-Split-Huge-CSVs-in-Sec...

Would love feedback from folks who deal with messy or oversized data. What features would you want in a CSV splitter?

I’ve just relaunched https://olllo.top with a brand-new design, logo, and overall presentation. Data.olllo is a no-code data analysis tool that runs on your desktop, aimed at helping users analyze large files quickly, privately, and without needing to write code.

What’s new: • Fresh branding and visual identity (logo, colors, layout) • Simplified homepage with clearer descriptions of what Data.olllo does • Improved demo experience and easier navigation • A focus on real-time, local data handling – no uploads required

You can try it directly on Windows (7 days trail after sign up), and I’d love to hear your feedback—whether about the design, the messaging, or the tool itself.

This is something I’ve been building personally, and I’m here to answer questions and gather thoughts.

Thanks for checking it out!

→ https://olllo.top

Data.olllo is the only desktop tool that lets you open, filter, and visualize massive CSV datasets—all offline, with no coding and no cloud. Break free from Excel’s limits and unlock true data power.

Open 100GB+ CSVs instantly—no RAM bottleneck Visualize & filter millions of rows in real time Convert CSV to HDF5 for blazing-fast analytics No coding, no cloud, no data limits 100% local: your data stays private

Have you ever struggled with a CSV that's just too big? Whether you're hitting Excel limits, uploading constraints, or simply want to work on a smaller piece of the puzzle — Data.olllo has you covered.

With just a few clicks, you can split any large CSV file by:

File Size — Define the number of files and let Data.olllo do the rest. Column Values — Automatically group and split the data based on any column (e.g., Region, Category, Date). Direct Split (No Load) — Instantly split a massive CSV by row count without opening it first, for maximum speed and minimal memory use.

Working with large datasets—50GB, 100GB, or more—is a serious challenge in most tools.

Excel can’t open files this size. Python scripts take time to load and debug. Even many "pro" data platforms get sluggish or crash outright.

That’s why Data.olllo was designed to open massive CSV and HDF5 files without breaking a sweat—up to 100GB and beyond.

I’m curious how others here approach data tasks: Do you prefer writing SQL directly, using Python/pandas, or working with no-code/low-code tools (like Tableau, Airtable, or internal tools)?

For context, I’m building Data.olllo, a desktop app for processing CSVs locally — no upload, just fast filtering, transforming, and exploring data with a spreadsheet-like UI.

What do you personally reach for first when cleaning or analyzing a dataset?

Thanks for the feedback! Data.olllo isn't Electron-based—it's built in Python with tkinter and custom tkinter, so the size mainly comes from the data libraries and embedded Python environment. I agree that keeping things lean is important, and I’m actively working on optimizing the package size further.

Appreciate the DuckDB comparison—great tool and definitely a benchmark worth learning from!

Absolutely—CSVs are still everywhere, especially for simple interchange between teams and tools. I designed Data.olllo with that in mind.

That said, I also plan to add support for Parquet and other formats soon—definitely agree it's gaining traction for larger, structured datasets.

Tad is a great tool—very clean and useful for quick exploration.

Data.olllo is focused more on local data processing, not just viewing—things like filtering, transforming, merging, and even running Python code (with AI assistance coming). It’s built for both small and large files with performance in mind, using many cores including Polars under the hood.

Also, good news: the macOS version is in the works and will be submitted to the Mac App Store soon!