Aurimas Griciūnas

When data agents fail, they often fail silently - giving confident-sounding answers that are wrong, and it can be hard to figure out what caused the failure. "Building and Evaluating Data Agents" is a new short course created with Snowflake and taught by Anupam Datta and Josh Reini. This course teaches you to build data agents with comprehensive evaluation built in. Skills you'll gain: - Build reliable LLM data agents using the Goal-Plan-Action framework and runtime evaluations that catch failures mid-execution - Use OpenTelemetry tracing and evaluation infrastructure to diagnose exactly where agents fail and systematically improve performance - Orchestrate multi-step workflows across web search, SQL, and document retrieval in LangGraph-based agents The result: visibility into every step of your agent's reasoning, so if something breaks, you have a systematic approach to fix it. Sign up to get started: https://lnkd.in/gVwKuCCQ

6,636

222 Comments

What I love about #ApacheIceberg is how many different workloads folks are running on the same copy of data. Graph queries are a really cool example. Instead of copying data into specialized graph databases, PuppyGraph lets you run graph traversals directly on your existing Iceberg tables. Pretty cool stuff. But here's the thing - to get truly great performance, you need to ensure your Iceberg tables are well maintained and optimized. That's where Ryft comes in. We make sure your Iceberg tables are automatically maintained and optimized, no matter what engine you are using. It was terrific to work with Zhenni Wu and Weimo Liu, to create this blog post on how you can use Ryft with PuppyGraph to run graph queries on a fully optimized Iceberg lake! https://lnkd.in/giwnwMuc #ApacheIceberg #GraphQueries

99

2 Comments

🚀 New in Docling: Structured Data Extraction from Documents! 🚀 We’ve just added a brand-new functionality: extraction of structured data from complex documents using free-form schemas 🤩 What does that mean? 🔹 You can now skip the conversion step — instead of turning documents into text or JSON first, Docling directly extracts the structured fields you care about. 🔹 The requested fields are defined in a free-form schema, so you can instantly align the extraction with the schemas of your own databases. 🔹 This makes it ideal for data-pipelines that don’t need full document conversion, but do need to populate structured databases from messy, unstructured documents. ✨ It’s: 1️⃣ Super simple to use (check out the code snippet 👉 https://lnkd.in/eK6vaMEe ) 2️⃣ 100% open-source and runs fully local – no API calls needed 🙌 3️⃣ Powered by cutting-edge models from NuMind (YC S22) 4️⃣ Perfect for data-pipelines where you need to populate structured databases from documents (think invoices, Curriculum Vitae, contracts, product datasheets, etc) 5️⃣ Currently focused on PDFs and images (PNG) — support for pure text coming soon! The example below shows how easy it is to define a schema and extract structured fields directly from an invoice. 👉 Try it out, break it, send us feedback, and if you like what we’re building, don’t forget to ⭐ the repo (https://lnkd.in/d4UT-6_2)! After a long and fruitful summer, the Docling team has been cooking up many new features — and this is just the beginning! 🌟 #opensource #AI #documentAI #docling #IBM #IBMResearch

1,382

83 Comments

Steps here👇the game has completely changed since 2022 - it’s not just leetcode anymore. Here’s what Viktoria Semaan shared that actually worked: ✅ Actually USE the platform (yes, she had to learn Databricks + Spark from scratch) ✅ Nail the coding assignment without ChatGPT crutches ✅ Come with CLEAR impact stories for leadership rounds ✅ Do your homework on every interviewer (LinkedIn stalking is real prep 😅) The technical bar is higher, but the human connection piece is everything. When you can tie your stories to their experience and show genuine interest in their work - that’s when you stand out. 2025 interviews hit different. Preparation is everything 💪 What’s the biggest interview challenge you’re facing right now? Drop it in the comments 👇 #techjobs #interviewtips

187

5 Comments

Is your team overthinking their data quality checks? I use a simple 3x2 matrix to think about which data quality checks to get started with. Overall, ➤ Every dataset needs declarative table-level checks. ➤ Most production datasets need record-level checks. ➤ ML-based checks are reserved for high-value or high-risk pipelines where silent anomalies are expensive. Too many checks add noise and slow teams down. Too few checks and bad data reach the business. Do not forget about ownership here: declarative checks typically sit with data stewards, while ML-based checks fall to engineering How do you decide the right level of coverage for your pipelines?

101

3 Comments