Insights, tutorials, and updates on workflow orchestration, durable execution, and AI agents from the Orkes team.
A step-by-step tutorial for building an Orkes Conductor worker: register a task definition, write the worker, and wire it into a complete workflow with branching.
What is a worker in Orkes Conductor? Learn how workers let you run your own code as a step in a workflow, how polling works, and when to write one.
Build an AI agent that turns UI screenshots into working React + TypeScript components. Drop in a design, get back clean, Tailwind-styled code in seconds.
Loop engineering is building a loop around your agent so the agent can reprompt itself without you needing to do it. but you still need to keep yourself in the picture.
Part one of a series on durable loops. Why agentic systems are loops, why naive loops fail as tasks get longer, and what makes a loop durable, grounded in a working Conductor OSS example.
Every time you copy workflow logic into a new workflow, you create a maintenance problem. Here's how the Sub Workflow task solves it.
Wiring an MCP server into an agent usually means scattering API keys across every process that runs it. How to connect MCP tools with Agentspan and keep the credentials on the server.
You have a workflow that needs to process a list of items and doing them one by one is too slow. Here's how the DYNAMIC_FORK task solves it.
You have a workflow that needs to take a different path based on a score, a status, or a region. The last thing you want to do is build that routing logic into your application code. Here's how the SWITCH task solves it.
Most workflows need to talk to the outside world at some point. Here's how Conductor's built-in HTTP task handles API calls, webhooks, and external service triggers without any custom code.
You have a workflow that needs to calculate a price, apply a business rule, or format a string. The last thing you want to do is deploy a whole service for that. Here's how the INLINE task solves it.
An agent crash shouldn't erase everything the run already did. How to make skills.md skill invocations durable, recoverable executions with Agentspan.
Stop writing custom workers just to transform JSON. Here's how Conductor's built-in JSON_JQ_TRANSFORM task handles reshaping, filtering, and merging in one expression.
Learn how conversational memory keeps AI agents coherent across follow-ups, corrections, tool calls, and durable sessions, with practical examples in Agentspan.
How conductor-skills teaches any AI coding agent—Claude Code, Cursor, Copilot—to create, run, and monitor Conductor workflows without you explaining a thing.
A practical introduction to semantic memory for AI agents: what should persist across sessions, how it differs from chat history, and how Agentspan exposes memory as agent infrastructure.
A practical system design guide for production AI agents: tools, state, integrations, human approval, recovery, observability, and deployment architecture.
Let’s build a customer success AI agent that can understand customer issues, gather context, and decide what to do next..
A walkthrough for treating prompts as first-class, testable artifacts before they ship inside your agents and workflows.
Swap run() for stream() and watch every reasoning step, tool call, and result land in real time — without giving up Agentspan's durability.
Build an issue triage bot in Agentspan with Strategy.ROUTER, where a dedicated classifier agent routes each issue to the right specialist — cheaper and more explicit than handoff.
The four building blocks behind every AI agent — LLMs, tools, loops, and memory — and how to connect them up into something durable.
Build a human-directed editorial workflow in Agentspan with Strategy.MANUAL, where you pick which agent runs next — writer, fact checker, or copy editor — every turn.
Build a multi-agent brainstorming session in Agentspan with Strategy.RANDOM, where agents are selected randomly each turn to generate diverse, unpredictable perspectives.
Build a code review debate in Agentspan with Strategy.ROUND_ROBIN, where specialist agents take turns, respond to each other, and converge on a verdict.
How you can use Claude Code and Conductor Skills to build all of your workflows
Build a support swarm in Agentspan using Strategy.SWARM so specialist agents can transfer work directly to each other.
A deep dive into a durable coding assistant REPL built with Agentspan, covering the agent loop, tools, event streaming, workflow message queues, signals, background processes, and resume.
A practical look at why durable execution, server-side state, and workflow observability matter when AI agents move from demos to production.
Use Agentspan `@tool` functions and `ToolContext.state` to compare SBOMs, apply review policy, and keep the full execution trail on the server.
Use Agentspan `api_tool()` to turn an existing OpenAPI spec into agent-callable operations, scope the API surface with `tool_names`, and inspect the resulting execution in the UI.
Why durable agent execution needs a runtime-owned instruction pointer, an intent ledger, and effectively-once side effects instead of a process-local loop.
Use an output guardrail in Agentspan to rewrite sensitive incident metadata before the workflow returns a customer-facing response.
Build a parallel code review pipeline with Agentspan using specialist agents that review the same code at the same time.
Use approval-required tools in Agentspan to pause a workflow for human review, then resume the same execution after approval.
Build a sequential AI support pipeline with Agentspan using specialist agents for classification, response drafting, and escalation.
Deploy a durable multi-agent workflow on Kubernetes with Agentspan, K3d, and a crash-safe orchestration model.
A practical introduction to AI agents, including loops, tools, memory, and how simple agents differ from more complex systems.
Build an incident-response agent that can investigate, pause for approval, and resume safely after crashes.
Why Agentspan exists, what it ships today, and how it brings durable execution to production AI agents.
How to install, configure, and use the Conductor CLI to manage workflows without leaving your terminal.
This support ticket triage automation workflow uses AI to read support emails, pull out important details, and send each ticket to the right team automatically.
How to ship new workflow versions without breaking in-flight executions.
A practical walkthrough for preventing duplicate side effects with idempotency patterns and Orkes Conductor.
Automate Tesla Powerwall export curtailment during negative Amber pricing using Orkes, accelerating solar payback with zero manual effort.
A technical walkthrough of MCP, why debugging MCP matters, how MCP Workbench is built to help, and a practical getting-started project you can validate end-to-end.
Learn how to orchestrate LangChain agents for production using Orkes Conductor with retries, human approval, and full observability.
Universal Commerce Protocol (UCP) explained: what it is, how it works, and why it matters for the future of AI shopping.
GPT-5.2 just shipped. Here’s how to plug it into Orkes Conductor and start running real agentic workflows (with guardrails, observability, and the flexibility to switch models anytime).
Company X employees are frustrated juggling support tickets. Learn how an agentic workflow can auto-triage issues so engineers get back to building.
Step-by-step tutorial to build an AI-powered loan risk assessment workflow in Orkes Conductor using OCR, LLM scoring, sub-workflows, and human review.
Step-by-step guide to building a Conductor workflow that monitors a website URL, evaluates health from HTTP status codes, and texts you via Twilio when it’s unhealthy.
Vector databases explained simply and how Conductor makes working with them simple and straightforward by integrating with popular vector databases like Pinecone, Weaviate, and Postgres.
Here’s a quick and easy rundown of how to use Orkes Conductor’s LLM embedding tasks to turn your text into vectors, store them in a database, and use them for things like semantic search, recommendations, and smarter routing in your workflows.
If you've ever struggled to connect tools like Slack and Supabase into one smooth workflow, this guide shows how to orchestrate the whole process, with Conductor quietly managing the hard parts.
Learn how to manually connect your Supabase database to Orkes Conductor using a JDBC integration, SQL queries, and simple workflow tasks.
Instantly connect Supabase with Orkes Conductor using a ready-made template. Automate data workflows and trigger actions in just two tasks.
Learn the magic behind AI agents and build your own ghostly chatbot; complete with memory, reasoning, and an external tool.
Learn what Role-Based Access Control (RBAC) is and how Orkes Conductor uses RBAC to secure workflows, prevent unauthorized access, and keep your automations safe.
Orchestration and choreography are two strategies for managing communication between services. Orchestration gives clarity and control, while choreography offers adaptability and independence.
A quick breakdown between the two agentic tasks powering your Conductor workflows. What they are and when to use each.
A quick and easy guide to connecting your NextJS app to your Orkes Conductor workflow. With code examples, as always.
Breaking big prompts into smaller ones makes workflows clearer, faster, and stronger. Here’s how to do it with Orkes Conductor.
Recap of fintech_devcon 2025: intimate, inspiring, and filled with community, connection, and a few surprise highlights.
What exactly is a human-in-the-loop and why does it matter for agentic AI? And what happens when humans and AI make decisions together in real time?
Securely integrate Orkes Conductor into enterprise environments with Proxy Support for HTTP and HTTP Poll tasks.
A concise playbook for running agentic workflows in production: guardrails, checkpoints, and the Conductor features that eliminate common performance bottlenecks.
Fraud doesn’t wait. Neither should your defenses.
Orchestration coordinates services, agents, and humans into automated flows, reducing fragmentation, tech debt, and operational inefficiencies.
Announcing the Conductor MCP Server — a lightweight API layer that gives AI agents real-time control over Conductor workflows.
SerDe tests ensure that data objects are correctly converted for interoperability. Learn how to automate test code creation using orchestration and AI.
A packed night of Mastering AI Agents and Agentic Workflows at SPIN Seattle. Top takeaways, fun moments, and a glimpse at what’s ahead with Orkes Conductor.
What is Model Context Protocol (MCP)? Learn how MCP simplifies AI integration by enabling seamless connections between agents and third-party tools
Discover more about orchestration, a powerful alternative to BPM tools for designing and automating business processes at scale.
Agentic AI systems can be implemented as agents or as workflows. Learn how they are different and which strategy to use.
Learn how to use task retries, timeouts, and system-level safeguards in Orkes Conductor to build resilient, fault-tolerant workflows.
Learn how to use timeouts, retries, and failure workflows in Orkes Conductor to build resilient, self-healing distributed systems.
Learn how to use workflow timeouts and failure workflows in Orkes Conductor to build resilient, SLA-compliant orchestrations that recover gracefully from failure.
Instantly convert legacy BPMN files into modern workflows with the new BPMN Importer in Orkes Conductor. No code, no rewrites—just fast, accurate orchestration.
Learn the importance of using control flow operators to declaratively build dynamic, responsive workflows without plumbing code.
Master branching and conditional flows in Orkes Conductor workflows using operators.
Add loops, waits, and human approvals to Orkes Conductor workflows with simple, structured operators.
Discover how to scale workflows in Orkes Conductor with Fork/Join, Dynamic Fork, SubWorkflow, and StartWorkflow operators for parallelism and reusability.
Solve manual claims processing inefficiencies with AI automation and orchestration. Get an in-depth technical breakdown for building with Conductor.
Learn how to build a fully automated technical interview workflow using Orkes Conductor. This step-by-step guide walks you through orchestrating LLM-powered interviews—from candidate input to feedback delivery.
Learn how to use built-in system tasks in Orkes Conductor for real-life use cases.
Use built-in system tasks to accelerate development time, enable low-latency execution, and ensure enterprise-grade availability in Conductor.
Use agentic workflows to add AI-driven dynamism in a governable and traceable manner. Try it with Orkes Conductor.
Learn how to orchestrate long-running APIs using Orkes Conductor, AWS Lambda, API Gateway, and DynamoDB with asynchronous invocation and polling.
Agentic workflows are an AI-driven processed where the sequence of tasks are dynamically executed with minimal human intervention to achieve a particular goal.
Using asynchronous workflows enables responsive and resilient systems even in the face of uncertainty.
Find out how agentic AI, combined with BOAT tools, will accelerate the world of business automation into the realm of intelligent, adaptive decision-making.
Learn how to automate email nurturing workflows with SendGrid webhooks and Orkes Conductor for improved lead conversion and personalized follow-ups.
Using BOAT software, implement the ten best practices that will help you master business process orchestration.
Discover why service uptime monitoring is crucial for businesses. Learn how Orkes Conductor helps ensure 99.999% uptime and prevent downtime with automated alerts.
Learn hands-on the best practices to improve RAG accuracy and response quality: contextual headers, semantic chunks, hybrid search, and reranking.
Learn how to build an AI document classification app using Orkes Conductor and LLMs to automate and streamline document sorting and improve efficiency.
The AI trends businesses should watch for in 2025: 1. AI redefining user-computer interactions 2. Evolving AI capabilities 3. Agentic AI to mature 4. AI governance to follow suit…
Conductor OSS has a brand new visualizer for its workflow definition and execution diagrams.
Learn about realtime API orchestration and other key strategies to reduce latency, streamline API workflows, and optimize application performance.
Experiment with more effective LLM prompts by leveraging model choice, prompt engineering tactics, and parameter tuning.
Get a guided tutorial on how to develop an AI application with minimal technical overhead using orchestration.
The Backend for Frontend (BFF) pattern customizes backend services for each frontend. Learn how to implement a BFF layer with Orkes Conductor.
Candid Conversations is a new video series exploring trends in workflow orchestration, AI, and tech’s impact on modern app development.
Change Data Capture (CDC) is a technique for identifying and capturing changes in a database and replicating them in real time to other systems.
Orkes is thrilled to announce Java Client v4, featuring significant design enhancements, performance improvements, and optimized dependencies.
Figure out if your monolith should be split into microservices and learn the best practices for planning, executing, and testing your migration project.
Debugging distributed systems is tricky due to observability or reproducibility issues. A key strategy lies in using tools that cut out time to track down the error source.
Orchestration is the process of coordinating distributed software components and systems so that they execute seamlessly as an automated, repeatable process.
Software architecture has evolved from mainframes and monoliths into a distributed network of cloud computing, API connectivity, AI, and microservices.
Orkes is officially part of Singapore's IMDA Spark Programme as of March 2024, allowing Orkes to facilitate deeper engagement with the local tech community and tech leaders.
In AI orchestration, a central platform coordinates interactions between AI components. Discover the different ways to use AI in your business applications.
Orkes Conductor offers integrations with Amazon Bedrock, providing greater flexibility for orchestrating workflows using AI.
Learn the fundamentals of prompt engineering, what it is, its significance in app development, and how to build LLM-powered applications effectively.
RAG (retrieval-augmented generation) improves LLM output with pre-fetched data from external sources, enabling developers to build semantic search features.
Conductor is an orchestration engine for creating complex flows across distributed components. Learn the differences between OSS and Orkes.
Dive into the latest developments in AI, API, and automation and discover how to leverage them for your business.
Explore the highlights from the AI Orchestration Meetup in Bengaluru, featuring insights on GenAI, AI orchestration, and implementation using Orkes Conductor.
Learn more about the technical implementation of durable systems and how Conductor delivers resilience with its decoupled infrastructure and redundancy.
Durable execution enables applications to continue running in face of failure. Find out how to implement it, especially for long-running processes.
Learn how to use Conductor-Webhook integration to automate slack greetings to new community members.
Find out how to upgrade EKS clusters using Conductor, with in-built failure handling, scheduling, and more.
Orkes’ Series A funding will power the next phase of growth for Conductor as the development platform of choice.
Conductor’s Python SDK has been refactored, providing support for AI orchestration, a better experience for creating dynamic workflow, and more.
Learn more about the happenings from Orkes in Jan 2024.
Announcing the initial release of Conductor OSS from Orkes.
Check out the latest December 2023 updates in Orkes.
Learn how to integrate your Conductor orchestration stack with Opsgenie to get incident alerts and swiftly take action.
Learn from our community on how to Conductor OSS to set up a Digital Public Goods (DPG) compliant orchestration layer in your applications.
Orkes will take stewardship of Conductor OSS in collaboration with the community and will remain committed to a developer-first experience.
Check out the latest November 2023 updates in Orkes.
Seamlessly integrate LLM capabilities and vector databases into applications using AI orchestration and native prompt engineering support.
Check out the latest October 2023 updates in Orkes.
Learn the four most important patterns for microservices: CQRS pattern, strangler pattern, publisher/subscriber pattern, and saga pattern.
Check out the latest September 2023 updates in Orkes.
Automate wildcard certificate deployment at scale with Conductor so that cluster access is always kept secure.
Check out the latest August 2023 updates in Orkes.
Learn how to control a running workflow by sending signals to the workflow execution.
Learn how to use tools like Prometheus and Grafana for monitoring Conductor workflows.
Check out the latest July 2023 updates in Orkes.
Find out how to implement a saga pattern in your applications using Orkes Conductor, an orchestration tool.
Discover the best tools, languages, and frameworks for accelerating the development cycle, from API testing to performance monitoring.
Learn how to create a Slack bot that sends reminders to post daily stand-up messages using Conductor-Webhook integration
Check out the latest June 2023 updates in Orkes.
Get Orkes’ recap of the highlights from the 2023 Microsoft Build conference and Gartner Application Innovation & Business Solutions Summit.
Check out the latest May 2023 updates in Orkes.
Learn how to build a subscription management system with Spring Boot and Orkes Conductor.
Get a recap of the highlights from Orkes Bangalore meetup in April 2023.
Discover how to use webhooks in Conductor workflow for a variety of use cases in microservice communications.
Distributed transactions can safely recover from failures using rollback mechanisms like the saga pattern.
Use Conductor, a workflow orchestration platform, to automate cloud infrastructure upgrades with no downtime.
Check out the latest April 2023 updates in Orkes.
Use Conductor to transition from a monolithic to a microservice-based architecture and to modernize your applications.
Transactional backends can be implemented using a database-oriented approach or a workflow-oriented approach.
Discover more about Conductor’s Scheduler feature and automate your workflows at custom intervals.
Check out the latest March 2023 updates in Orkes.
Learn the differences between orchestration and choreography, and the advantages and drawbacks of each approach.
Conductor is a workflow engine, which can help automate routine tasks, enable faster application deployment, and more.
Writing workflows with code unlocks greater flexibility in creating dynamic workflows on the go. Try out Conductor’s SDKs today.
Check out the latest February 2023 updates in Orkes.
Discover how to use LLMs, like ChatGPT, as a coding co-pilot to generate Conductor workflows as code.
Learn how to automate a long-running user subscription workflow using Conductor, an orchestration platform.
Check out the latest January 2023 updates in Orkes.
Learn how to build a checkout application using Next.js and a workflow as code approach in Conductor.
Based on benchmark tests, Conductor’s workload performance demonstrates consistent throughput across task execution, data processing, and more.
Use Conductor to build transaction dispute workflows and get native support for retries, scaling, debugging, and more.
Explore Orkes updates and highlights from December 2022.
Get the key highlights from Orkes's participation at DevFests in 2022.
Check out the latest November 2022 updates in Orkes.
Orkes’ first hackathon, the Orkes Hack, was held online and participants competed to create the most creative workflow in Conductor.
Orkes has attained SOC 2 Type 2 compliance in our continued endeavor to provide the highest standards in data security for Orkes Cloud.
Orkes Conductor has webhook integrations available for seamless integration with third-party applications.
Learn how to build a food delivery application using TypeScript and a workflow as code approach in Conductor.
Orkes Conductor supports the use of secrets to securely send sensitive parameters in workflows.
Discover how to implement KYC (Know Your Customer) checks for high-value transactions using Conductor.
Learn how to reuse tasks and workflows in Conductor, with reusable tasks, sub workflows, dynamic tasks, do/while operators, and more.
Learn how to break up a monolithic application into microservice-based architecture for greater scalability and ease of deployment.
Monolithic architecture creates a tightly coupled codebase that takes longer to update and deploy, which can be resolved by breaking it up into microservices.
Using multiple languages in a microservice architecture provides greater flexibility, but may require more maintenance.
Learn how to build a Conductor workflow for data processing, which follows the three-phase ETL (extract, transform, load) process.
Announcing the new open-source repository for Orkes Conductor.
Polyglot microservice orchestration enables developers to use multiple programming languages for various purposes in a single workflow.
Try using Conductor to run tasks and workers in Clojure.
Workflows are like recipes: a series of tasks that must be completed in order, with varying degrees of complexity.
Learn the benefits and limitations of a microservice architecture, with an example use case in virtual goods.
Hybrid cloud architecture makes it easier to manage, scale, and migrate your infrastructure while reducing operation costs.
Orkes has launched new Conductor SDKs for C# and Clojure and made major improvements to Golang and Python SDKs.
Microservice orchestration helps teams to build and scale applications swiftly without creating a distributed monolith.
Conductor is an orchestration platform that connects distributed microservices so data flows without needing to build additional infrastructure.
Announcing Conductor Playground, a full-featured sandbox environment to try Orkes Conductor for free.
See workflow orchestration in action with Conductor, using a loan banking workflow example.
Announcing Orkes Cloud, a hosting solution that manages your Conductor deployment and infrastructure for you.
Get a recap of Orkes’ Feb 2022 developer meetup, where invited speakers share how they use Conductor in production at their companies.
Learn how to modularize and re-use workflows in Conductor using sub workflows.
Learn how to use Conductor’s dynamic forks, which creates the appropriate number of parallel processes at runtime.
Learn how to use Conductor to build an image processing workflow that can automatically create multiple sizes and formats for the same image.
Learn how Conductor handles retries and timeouts with its built-in error-handling functionalities.
Learn how to use Conductor to build an image processing workflow that can automatically complete tasks like resizing or format conversion.
Orchestration provides a systematic approach to coordinating microservices and offers more visibility into process flows compared to choreography.
Find out how Conductor got its start as an open-source orchestration engine, originally built by Netflix to solve infrastructure issues in microservices.