ElevenAgents Architect for live voice agents
ElevenAgents Architect drafts and tests fixes for a voice agent in production, so your engineers review changes instead of hunting for them.
Raja Vemuri · 5 minAI
Blog
Field notes from the engineers.
Our senior engineers build data platforms, AI agents and long-running systems alongside our clients' own teams, and they write up what happened when those systems met production, including what broke and what we'd do the same way again. If an AI pilot of yours never got past the demo, or a data project is waiting on headcount, you'll find the same problems here and how our engineers dealt with them.
Margin notes from the Databricks Field Guide
All margin notesShort takes on what Databricks just shipped and what we read into it, published on the Field Guide.
- Oct 13, 2026 · TechFabric
Databricks Private Network Gateway: serverless into your network
Private Network Gateway gives Databricks serverless compute one connection into your own network. What it reaches, what it costs and where to pilot it.
Read on the Field Guide - Oct 12, 2026 · TechFabric
Databricks Feature Views are GA: what changes for feature code
Databricks Feature Views reached general availability on 12 October 2026 for batch, streaming and request-time data. What to check before adopting them.
Read on the Field Guide - Sep 13, 2026 · TechFabric
Databricks LTAP Direct Writes in beta: faster synced table loads
LTAP Direct Writes, in beta since 25 August 2026, loads synced tables by writing to storage instead of through compute. The first LTAP claim a team can time.
Read on the Field Guide
99 posts
Databricks Data Quality Gates: From Failed Checks to Blocked Releases
Turn Databricks quality checks into release gates. Validate a fixed candidate, block publication on failure, and preserve the previous approved data.
Securing long-term agent memory on Databricks
Databricks agent memory persists across conversations, so a wrong fact outlives the chat that created it. How we'd govern what enters a memory store.
Ivan Vydrin · 10 minDatabricksAICustomer-held encryption keys on Azure Postgres Flexible Server
Cross-tenant customer-managed keys let a customer hold the key to their data in your Azure Postgres app. What that costs your team, and when to build it.
When Azure PostgreSQL needs a backup vault
Azure's built-in PostgreSQL backups stop at 35 days and outlive a deleted server by five. Which production databases need a Backup vault, and what it costs.
Ivan Vydrin · 9 minAzureCloudAzure Databricks private network gateway for serverless compute
Azure Databricks private network gateway gives serverless one path into your VNet and on-premises network. How we'd pilot it, and what we'd wait on.
Andrey Taran · 7 minDatabricksAzureGemini 4 Argon pricing for AI agents
Argon's token price doubles on December 1. Test it against your current model by replaying real past tasks, judging blind and comparing cost per finished task.
Sushruth Aeluguri · 11 minAIGoverning AI agents with Databricks and Temporal
How Unity Catalog and Unity Gateway limit what an AI agent can do, how metric views and Genie Ontology define its metrics, and why we run agents on Temporal.
MCP Events let an agent react without being asked. Here's how we'd make one survive a restart
MCP Events let a customer's agent subscribe to your platform and act the moment something changes. Where Temporal and Databricks belong behind it.
Agent 365 plus API Management: MCP tool governance you can actually enforce
Microsoft's Agent 365 and Azure API Management preview let you decide, in one place, which company tools your AI agents can use. Who should care, and the cost.
Ihor Seleznov · 8 minAIAzureA DuckDB view-only catalog over Delta Lake
DuckDB can ship a few hundred KB of view definitions over a Delta Lake. Where we'd use it on client work, and the Unity Catalog line we would not cross.
Muse Wonder: Meta's frontier models for agents
Meta's Muse Wonder claims to pause before risky actions and remember context across sessions. What that would change for approval gates on production agents.
Preetham Reddy · 8 minAINewsResponsive joins Databricks: what Lakebase gains
Responsive, the SlateDB team, is joining Databricks. Which Lakebase limits that should lift, judged against the ceilings we design around in client work.
OpenCode 2.0 turns the open harness into a service. Here's how we'd use it to cut the token bill
Your coding agents' token bill keeps growing? What OpenCode 2.0 changes, and the config we'd commit to route routine work to cheaper open-weight models.
Preetham Reddy · 8 minAIKnowledge ShareBC data mirroring to Fabric: what changes this quarter
Business Central can now copy any table into Microsoft Fabric for reporting, with no export or pipeline to maintain. Who should move, and where it stops.
Cloudflare's new web search and web fetch tools for AI agents are impressive. Here's why we switched
Why our AI agents switched to Cloudflare's web search and web fetch the day after launch, and what it fixed in our fact-check.
Preetham Reddy · 7 minAICloudA shared, network-safe web reader for AI agents
If your AI agents read web pages, a link somebody else chose can reach your internal network. What that hole looks like, how to close it, and what's still open.
Preetham Reddy · 7 minAICloudProgrammatic SEO with AI agents, the white-hat way
How to scale content with AI after Google's September 2026 spam update without doorway pages or scaled content abuse, and what kept our city pages indexed.
Microsoft Execution Containers: contain your AI agents
MXC is Microsoft's policy-driven sandbox for AI agents, now GA. What it enforces, the JSON policy, the three modes, and where it stops short.
TechFabric · 6 minAINewsAI decision models compared: Jev, Clef, Kev, Microsoft
Which AI model should make your software's quick calls, like routing a ticket or approving a draft? Jev, Clef, Kev and Microsoft-Decision-1 compared.
Preetham Reddy · 8 minAIKnowledge ShareSkillFS: governing agent skills with a file system
SkillFS is a FUSE file system that controls which agent skills are visible, which are trusted and what happens when one goes bad.
TechFabric · 8 minAIKnowledge ShareBC29: multiple excise duties per item in Business Central
Business Central 2026 wave 2 lets one item carry several excise duty configurations. What changed, how to set it up, and where it stops.
Funke: native HL7v2 parsing on Databricks, explained
Databricks open-sourced Funke, a Python parser that lands HL7v2 messages in Unity Catalog with the full segment hierarchy intact.
Embedded analytics on Databricks, part 2
Five datasets per board meant five copies of the same query on every filter click. The rule we use now, and the three places it breaks.
Replit on Databricks: governed apps with Lakebase Postgres
Replit Agent now builds apps on live Databricks data and deploys them as Databricks Apps with an auto-provisioned Lakebase Postgres database.
Supermemory is where we'd keep what an AI agent learns about a user, not the facts a business owns
How TechFabric would use Supermemory for agent memory on client work. Which memories belong in it, what decides the bill, and the defaults we'd change.
Preetham Reddy · 8 minAIKnowledge ShareDatabricks bought Row Zero: spreadsheets on live data
What the Row Zero acquisition means for finance teams living on exports from Business Central and a CRM, and what is announced but not yet shipped.
Clef, Clef-flash and Clef-omni compared
Which of Cloudflare's three Clef decision models should make a call your software repeats all day, and how we'd build a manufacturer's warranty-claim review.
Preetham Reddy · 8 minAIKnowledge ShareDatabricks Lakebase: managed Postgres next to your lakehouse
What Databricks Lakebase is, how branching, point-in-time restore and synced tables work, who uses it, and when to keep your existing Postgres.
BC29 retained DEV extensions: VS Code apps survive updates
Business Central 2026 wave 2 keeps DEV extensions installed through an environment update. What changes for sandboxes, UAT and your release process.
Business Central BC29: turning indexes on and off from AL
BC29 lets AL developers enable and disable database indexes in code. What that changes, how a team would use it, and where it does not fit.
Lakebase database branching: a database per coding agent
Databricks Lakebase branches a Postgres database in under a second. Here is how that gives every coding agent and every PR its own isolated database.
Voice agents on Databricks for auto refinance
How a compliant outbound voice agent for auto refinance is built on Databricks. Consent gating, feature lookups, the agent runtime and what to measure.
Migrating to Databricks without a cutover spreadsheet
How we move Synapse and Snowflake estates onto Databricks with Lakebridge and Airlift. One manufacturing migration took six weeks instead of six months.
GPT-6: choosing Astra, Sol or Luna and setting effort
How the GPT-6 family splits by cost and reasoning effort, the API rules that changed, what it costs per million tokens, and how to run it on Databricks.
TechFabric · 8 minDatabricksAIMeta's ads MCP server on Databricks: what it changes
Meta's ads MCP server is in Databricks Marketplace. What it lets an agent do, how Unity Gateway and Meta rules fence it, and where it does not fit.
TechFabric · 9 minDatabricksAIBusiness Central MCP server: query data with no API page
Business Central agents could only read a few standard tables without custom development. Version 29.0 lets them query the rest. How it works, where it stops.
TechFabric · 10 minDynamics 365AIDatabricksLakebase vs Snowflake Postgres: branching is the difference
Choosing a managed Postgres for apps beside your data platform? Lakebase against Snowflake Postgres on branching, scale to zero and governance. Lakebase wins.
Starting our SOC 2 journey: what we found before the audit
Customers asking for a SOC 2 report? Preparing for SOC 2 Type II, we found the controls were already running and proving them was the work.
Unity Catalog for AI agents in practice: four FHIR scenarios
Agent identity decides what your row filters do. Four scenarios from a FHIR lakehouse on Databricks, with the performance tradeoffs of each.
Decision models in auto refinance: where Jev and Clef fit
Lenders kick back refinance packages over a mismatched odometer or an expired payoff. Where decision models like Jev and Clef catch it, logged on Databricks.
Preetham Reddy · 8 minAIDatabricksBusiness Central reporting: built-in, Power BI or lakehouse
Three ways to report on Dynamics 365 Business Central, what each one is good for, the limits Microsoft documents, and how to tell which one you need.
An AI governance framework is four answers you can query
The EU moved its high-risk AI deadline to December 2027. What NIST AI RMF, ISO 42001 and the EU AI Act ask, mapped to controls you can run on Databricks.
LakeWright.NET carries a SaaS tenant boundary into Databricks
An Apache-2.0 .NET kit that resolves the customer tenant before a request reaches Databricks SQL, Jobs, AI/BI, Genie or Model Serving.
Embedded analytics on Databricks, part 1
One ticket, one platform, and the cheapest first step. Moving a B2B product's client reporting onto Databricks, starting with Lakehouse Federation.
Dynamics NAV support ends in January 2028
Microsoft's extended support for Dynamics NAV 2018 ends on 12 January 2028. What moving to Business Central involves, and why the warehouse floor sets the size.
TechFabric · 4 minWhat We DoDynamics 365TechFabric engineer named a winner in Databricks' Genie-Powered App Challenge
Prove It, by TechFabric engineer Ivan Vydrin, won in the Creativity & Originality track of Databricks' global Genie-Powered App Challenge, judged by Databricks.
TechFabric · 2 minDatabricksNewsFabric or Databricks is the wrong question on Azure
Microsoft and Databricks shipped the interop this year. For teams already on Azure the question is which platform holds the tables and which one reads them.
The lock-in argument for leaving Snowflake is over
Snowflake and Databricks now read and write each other's Iceberg tables. If a vendor is still selling you a migration on lock-in, they are selling you 2023.
Export to Data Lake stops on 30 November
Export to Data Lake stops permanently on 30 November 2026. Microsoft names two replacements for Dynamics 365 teams. There is a third answer.
Nobody wanted to run the migration on a Friday
Staging has ten thousand rows, production has forty million, and every migration is a guess. What Lakebase branching does to a team's calendar.
Paying frontier prices to rename a variable
Databricks' smart routing sends routine coding-agent work to cheaper models and keeps the agents' traces in Unity Catalog. What it saves, and what to check.
Preetham Reddy · 3 minDatabricksAIThe pipeline was fine. The data was three days old.
Every job green, nothing failing, and the numbers quietly wrong because a source stopped sending. Freshness, volume and schema checks in Unity Catalog.
Conflicting metric definitions in Databricks Genie
Genie fails when two teams mean different things by the same word. Genie Accuracy hardens the semantic layer. TechFabric Experiments keeps the suite running.
Four thousand objects and nobody would name the exclusions
Warehouse migrations slip because the exclusions were never written down. The Migration Readiness Sprint inventories the estate before anyone picks a date.
The cost dashboard said the workspace was fine
A Databricks cost dashboard shows spend. A two-week Health Check finds the spend that bought nothing, from all-purpose clusters to orphaned schedules.
Unity Catalog before the first pipeline
Lakehouse pilots skip grants and then cannot be handed to security. The Lakehouse Launchpad puts Unity Catalog first, with TechFabric Airlift underneath.
Nobody could tell whether the agent was any good
A team shipped an agent in nine days and argued for two months about whether its answers were right. Write down what a good answer is first.
Databricks Repos for QA test automation workflows
Moving QA off ad-hoc notebooks and onto version-controlled test automation with Databricks Repos. Repo structure, modular tests, and how a team works in one.
Using PyTest for Data Quality Checks Directly within Databricks
Writing PyTest data quality checks inside Databricks notebooks. Bronze-to-silver validation, edge cases, fixtures and error handling, with examples.
Governance You Can Actually Ship
AI governance fails when it lives in documents instead of infrastructure. Compliance comes from systems that enforce the rules while the model is running.
TechFabric · 8 minField NotesDatabricksThe Agent That Deleted Production
Enterprise software fails when IT mandates tools people do not want. Adoption comes from pull, and the same rule now applies to agents.
TechFabric · 9 minField NotesThe Most Expensive Software Nobody Uses
Companies spend millions on enterprise AI that employees barely touch while shadow AI spreads. The problem is the tools, not the people ignoring them.
TechFabric · 9 minField NotesThe Forward-Deployed Playbook
AI changed build versus buy. Now it is changing delivery. Small teams of senior engineers embedded in your operations, shipping in weeks rather than months.
TechFabric · 12 minField NotesWhat If Agile Was the Detour?
Agile answered expensive execution, not planning. As AI collapses build time, deep upfront specification plus fast execution ships in weeks.
TechFabric · 12 minField NotesAgentic Systems Without the Chaos: Building AI That Stays on the Rails
The engineering patterns that keep an autonomous system reliable in production. Guardrails, observability, circuit breakers, and where each one belongs.
TechFabric · 15 minField NotesThe Build vs. Buy Decision for Enterprise AI (When Off-the-Shelf Fails You)
Faster models and larger context windows make a custom AI build viable in weeks, and data sovereignty exposes vendor limits. When to buy, build, or do both.
TechFabric · 9 minField NotesHow to reach a new market without rebuilding your core platform
When onboarding smaller customers pulls engineers off the roadmap, build a self-serve way in and leave the core platform alone. How to find where the cost sits.
TechFabric · 5 minField NotesWhat Teams Get Wrong About Outages and Operational Risk
Cloudflare, AWS and Azure all went down in late 2025. What their postmortems show, and the restore drills that decide how long your own outage lasts.
DASF - Databricks AI Security Framework
DASF stands for Databricks AI Security Framework. Version 3.0 maps 97 AI risks to 73 controls. What it covers, and how to enforce those controls at runtime.
Preetham Reddy · 2 minDatabricksUnity Catalog for enterprise data governance
Unity Catalog is the permission boundary we refuse to route around. A Health Check finds coverage that stops halfway, and agents inherit a person's grants.
Preetham Reddy · 3 minDatabricksDatabricks AI/BI dashboards and Genie
What Databricks AI/BI is: dashboards plus Genie, which answers plain-English questions with SQL. What changed in 2026, accuracy, and cost.
Preetham Reddy · 7 minDatabricksAIFrom data warehouses to data intelligence
How data platforms went from warehouses to lakehouses to data intelligence, and the four platform decisions that are hard to undo later.
Preetham Reddy · 6 minDatabricksAIMigrating from Legacy Data Warehouses to Databricks Data & AI Platform
Warehouse migrations slip on scope, not conversion. Two weeks on TechFabric Airlift inventories the estate, agrees exclusions and hands you a wave plan.
Preetham Reddy · 3 minDatabricksRethink AI: Building a Horizontal Layer for Enterprise-Wide Transformation
Horizontal AI versus vertical AI. Why businesses need a shared AI layer across their systems, what goes in it, and when a single-app copilot still wins.
Using Application Insights as a Sink for logging in ASP.NET Core
ASP.NET Core has very extensible logging interface. I provides an ILogger interface along with few default implementations that can be used to log data.
TechFabric · 3 minKnowledge ShareIntroducing Azure AI Foundry: A CTO's Guide to Starting and Scaling AI
Azure AI Foundry is now Microsoft Foundry. What it is, how it relates to Azure, and the day-one decisions a CTO makes on projects, data location and cost.
3 Ways To Run Automated Tests On Azure DevOps
Catch bugs before production by running tests automatically in Azure DevOps. Unit tests on CI, Selenium on release, Test Plans on demand, with the YAML.
Ihor Seleznov · 24 minKnowledge ShareAzureSetting Up Production Ready Infrastructure For Microservices
Service Fabric is Microsoft's microservices orchestrator. Service discovery, fault tolerance and containers, set up for production rather than a demo.
Using Traefik Reverse Proxy For Securing Microservices On Azure Service Fabric
Securing microservices on Azure Service Fabric with a Traefik reverse proxy. Service Fabric is Microsoft's answer to Docker Swarm and Kubernetes.
Preetham Reddy · 15 minKnowledge ShareAzureMulti-tenant AI infrastructure in the cloud with Temporal
Some tenants want a cheap shared setup, others an isolated one. How we keep LLM costs, vendor swaps and per-tenant deployments manageable with Temporal.
Durable RAG with Temporal and Chainlit
When one question to an AI assistant sets off several tool calls, any can fail halfway. How we made a Chainlit RAG chatbot durable with Temporal, with the code.
How We Created A Multi-Tenant, Multi-Cloud, & Multi-Model AI Platform with Temporal, Fiber Copilot
How we designed Fiber Copilot to stand up chatbots and copilots on your own data, in your own cloud, with the parser, vector store and model you choose.
Off the shelf vs. custom software: selecting the optimal CRM for your business
Off-the-shelf or custom CRM? Buy when your sales process is standard. Build when the workflow is your product. Most firms do best buying and extending.
TechFabric · 5 minCloudDynamics 365 CRM: Transforming Your Sales Quote Process
The Dynamics 365 quote process state by state, with the product catalog it needs. Templates and PDFs, approvals in Power Automate, and quote to order.
What Every CTO Needs To Know About Databricks
What Databricks is, what changed between 2024 and 2026, how the bill works and how to read it, and the governance and lock-in questions to settle first.
Power Automate and the small automation requests IT gets every week
Small automation requests swamping your IT team? How Power Automate takes them off engineers, and what to check first, from licences to AI Builder credits.
Karan Sharma · 7 minWhat We DoBreaking Down Silos: Integrating AI Across Business Functions
The AI worth building usually needs two departments' data. How Amazon and KLM joined theirs to cut waste, and how to find the same opportunity in your company.
Leo Oliemans · 5 minAIA wide configuration for running Cypress in GitHub Actions
How to make one test suite check every release in each environment, browser and screen size your customers use, from GitHub Actions, without rewrites.
Ihor Seleznov · 15 minKnowledge ShareDynamics 365 CRM - Why Are So Many Companies Migrating To It?
Why companies migrate to Dynamics 365 CRM in 2026, from on-premises Dynamics CRM, Salesforce and ERP-only estates. The dates, the prices and the route.
Hiring Databricks Specialists: 12 Expert Insights You Need to Get it Right
Twelve things to get right when hiring Databricks specialists in-house, from people who have built the teams and know what they screen for.
How Temporal Has Transformed My Development Approach
Retries, idempotency and state across services used to take more time than the business logic. A senior developer on what changed after adopting Temporal.
Mastering the Art of Prompt Engineering for Generative AI
A prompt can follow every rule and still get a confident, wrong answer from generative AI. One real prompt taken apart, fixed and tested before users see it.
Leo Oliemans · 7 minAIRobotic Automation: The Future Of Supply Chain Optimization
Logistics robots are the largest category of service robot. What they need from your ERP and warehouse system, what Dynamics 365 offers, and what to fix first.
Leo Oliemans · 5 minCloudTemporal: Airlines Booking Demo Application
Weighing Temporal for workflows that must survive failures in production? What our engineers learned modelling ticket purchases, returns and delayed flights.
The effects of Cypress component testing on your React app
Cypress component testing mounts one React component in a real browser. A working setup on Vite, tests with data-cy selectors and spies, and when to use it.
What durable execution is, and how Temporal's framework delivers it
Durable execution for teams whose order, payment or onboarding flows break halfway. How Temporal records every step and survives crashes, and where it stops.
Transform Your Data Visibility, Transform Your Organization
How organizations improve data visibility. Report off the database, add a read replica or build a warehouse, then use lineage to see how data changed.
TechFabric · 7 minWhat We DoUI/UX at scale: it's time to move to design systems
Why an app's interface slows the team down as it grows, and how a design system keeps one change one edit, with tokens, shared components and CI checks.
Sam Salima · 6 minDesign