Senior Software Engineer

Previewing Interrupt 2026: agents at enterprise scale. This year, Langchain is doing it again. Interrupt 2026 is May 13-14 at The Midway in San Francisco, and the lineup, the format, and the scale have all leveled up. Last May, 800 of you came to The Midway in San Francisco for the inaugural Interrupt conference. Teams from Cisco, Uber, J.P. Morgan, Replit, LinkedIn, and BlackRock got on stage and told the truth about what it actually takes to put agents in production. Langchain launched LangSmith Deployment, shipped a redesigned LangSmith Studio, and rolled out new observability tools in LangSmith. If you were there, you know the energy. If you weren't, here's a taste of what you missed: This year, Langchain is doing it again. Interrupt 2026 is May 13-14 at The Midway in San Francisco, and the lineup, the format, and the scale have all leveled up. 2026 is about agents at enterprise scale. Last year's question was "can agents work in production?" The answer, across dozens of talks, was a definitive yes. This year's question is different: how do you make them work at enterprise scale - and what does the team, the tooling, and the infrastructure look like when agents aren't a proof of concept anymore? Interrupt 2026 is about the how. How are the largest companies in the world building agent platforms? How are they evaluating performance when the stakes are high? How are they structuring teams around agent engineering as a discipline? And how is the ecosystem - from model providers to infrastructure - evolving to support what comes next? Keynotes and fireside chats with. Harrison Chase, Co-founder and CEO of LangChain, will open each day of Interrupt with a keynote on what Langchain has learned from working with thousands of teams shipping agents over the past year, where its products are headed, and predictions about the industry. Andrew Ng, founder of DeepLearning.AI and one of the most influential voices in AI, will share his view on what's coming next for agents, and what it means for the developers and teams building them today. Chirantan "CJ" Desai, CEO of MongoDB, will sit down with Harrison for a fireside chat on how the world's largest enterprises are building with agents, and what the data layer looks like when agents move from experiments to production systems. Aaron Levie, Co-founder and CEO of Box, the intelligent content management platform he launched in 2005 is a vocal advocate for AI-driven enterprise software and frequently writes and speaks on how organizations can use AI agents to transform workflows. What you'll hear from the stage. Langchain is bringing teams who are deep in production and running agents at real scale with real consequences. Here's a preview of what's on the agenda: Lyft is talking about evals and how they are building evals around their specific product policies, user flows, and edge cases with LangSmith. Nick Ung from Lyft's Safety and Customer Care team will walk through how they built an evaluation system that actually tells them whether their agents are working, and how they close the feedback loop between failed traces, their ops team, and engineering. Apple is sharing how they built a low-code agent platform serving 15,000+ employees. Their team rethought how LangGraph constructs graphs at runtime to support dynamic, low-code agent building at a scale that required rearchitecting assumptions about graph construction, caching, and context management. LinkedIn is presenting a solution to one of the biggest problems today: recruiting. Recruiting is one of the most time-intensive workflows in any organization - especially for small and mid-size businesses without dedicated hiring teams. LinkedIn's engineering team tackled this head-on by building an AI recruiting agent with LangSmith and LangGraph. Now thanks to their recruiting agent, the team is hiring 10x faster. You'll also hear production stories from the world's largest enterprises including Toyota, LATAM Airlines, and Honeywell, along with tech-native companies including Coinbase, Chime, Rippling, monday.com, and Clay. Beyond the talks. Interrupt hosts two full days designed around learning, building, and connecting. AMAs with product leaders. Sit down with its engineers building LangSmith, Deep Agents, LangGraph, and LangChain. Ask them anything - about the roadmap, about your architecture, about the problem you've been stuck on for weeks. Last year's product announcements came out of conversations exactly like these, and Langchain expect this year to be no different. Demo stations. Get hands-on with the latest across the LangSmith platform. The demo area spans the entire front patio and serves as the central hub of the conference, a place to see what's new, try things out, and talk to the engineers who built them. Workshops. Go deeper with hands-on sessions led by LangChain engineers. Langchain'll cover topics like building Deep Agents, as well as improving agents using LangSmith Align Evals and Insights Agent. These are designed to be practical. Bring your laptop and leave with tactics you can actually use. Time with speakers. One of the best parts of last year was the hallway conversations. This year Langchain has built even more space for that. You'll be able to meet with many speakers after their talks at its Ask Me Anything booth. Get your ticket. Interrupt 2026. May 13-14. The Midway, San Francisco. Langchain sold out last year and expect to again.

Anthropic is no longer a model company. Claude Managed Agents quietly redraws the competitive map for every AI infrastructure vendor. * | Anthropic launched Claude Managed Agents, shifting from selling model inference to selling a full agent platform. * | The move puts Anthropic in direct competition with AWS Bedrock, OpenAI Assistants, LangChain, and the agent framework ecosystem. * | Enterprise switching costs rise dramatically when agent state, tooling, and operations are locked into Anthropic's infrastructure. * | The unanswered question - who is liable when a managed agent causes real-world harm - will determine enterprise adoption velocity. Jessica Yan, a product lead at Anthropic, posted on LinkedIn yesterday to announce the public beta of Claude Managed Agents. It is worth reading carefully because it quietly describes the most consequential strategic shift at Anthropic in 2026. "You can now raise the ceiling of agent execution AND launch faster using our stateful APIs, performance-optimized harness, scalable infra, and rich developer tools." - Jessica Yan, Product at Anthropic Read those four capabilities in order. Stateful APIs. Performance-optimized harness. Scalable infrastructure. Rich developer tools. This is not a model release. This is not a feature expansion. This is Anthropic announcing that it is in the agent platform business - and by extension, in direct competition with AWS Bedrock, Google Vertex AI Agent Builder, OpenAI Assistants API, LangChain, LangGraph, CrewAI, Dust, and every other piece of infrastructure currently hosting agent workloads. Until Monday, building an agent on Claude meant you handled the infrastructure. Starting yesterday, Anthropic handles it for you. That is a business model change, not a feature launch. What actually shifted. Anthropic's business on Monday was selling inference. You bought API access to Claude, you handled state management, you wrote the orchestration, you built the monitoring, you scaled the infrastructure, you owned the developer experience. The margin was inference margin. The customer was anyone running a workload. Anthropic's business on Tuesday is selling a platform. You get Claude and the infrastructure to run Claude-powered agents in production. Anthropic captures more of the value chain. The margin is platform margin. The customer is the developer building agent products. Platform margins are higher than inference margins - that's the obvious part. The less-obvious part is stickiness. An enterprise that builds its agent on Claude Managed Agents cannot easily port that agent to a competing model. State, tooling, operational patterns, and incident history all get locked into Anthropic's infrastructure. Switching costs go up dramatically the moment a team's agent is running on Anthropic's harness. This is the move the cloud providers have been waiting to see. It's also the move they've been dreading. Who gets structurally worse this week. The hyperscaler Claude distribution path. AWS Bedrock and Google Vertex host Claude for enterprises that don't want to buy from Anthropic directly. Their value proposition is compliance, existing procurement relationships, and vendor consolidation. All three are real. None beat "the people who built Claude are also running your agent infrastructure for Claude." Every enterprise that was going to run Claude agents through Bedrock or Vertex now has a reason to evaluate going straight to Anthropic instead. The agent framework startups. LangChain, CrewAI, LangGraph, Dust, and a long list of others built their businesses on being the orchestration layer above multiple model providers. Their pitch was: don't lock into one LLM; build with its framework; switch models when you need to. That pitch just got harder. Anthropic can now offer deeper integration, better performance tuning, and direct first-party support for Claude-based agents than any third-party framework can match. The frameworks will reposition around multi-model interoperability. That's a harder sell than "we're the best way to build agents." OpenAI's Assistants API. OpenAI built Assistants to keep enterprise developers inside the OpenAI ecosystem. They will now have to respond to every Anthropic Managed Agents capability with an equivalent, while also fighting on ChatGPT Enterprise and the foundation model benchmark treadmill. OpenAI's response will come fast. It will also be reactive, not strategic. Who wins. Anthropic, obviously. They just expanded their addressable market from "developers buying model access" to "developers building agent products." That's a much larger number, at higher margins, with stickier customers. The subtler winner is any enterprise that was paralyzed on build-versus-buy for its agent infrastructure. Managed Agents doesn't eliminate the buy-side risk, but it gives risk-averse buyers a credible vendor-backed option they didn't have on Monday. Expect the number of enterprises that move from "planning an agent platform strategy" to "piloting Anthropic Managed Agents" over the next 90 days to be larger than most analysts expect. The question nobody in the coverage is asking. Here is what's missing from every take this week: when a Claude Managed Agent takes a real-world action that causes a real-world problem, who is responsible? The developer who wrote the agent? The enterprise that deployed it? Anthropic, whose platform is managing the state and executing the action? That question is not answered in Yan's LinkedIn post. It is probably not answered in Anthropic's initial documentation. It will be the first thing every enterprise general counsel asks before signing a contract, and it will be the single variable that determines whether Managed Agents gets enterprise traction or stays a developer tool. Anthropic has two ways to handle this. They can write a managed services agreement that places all liability on the customer. That's legally clean and will scare off exactly the enterprises most likely to pay platform prices. Or they can accept operational responsibility for the agents running on their platform. That solves the trust problem and fundamentally changes Anthropic's risk profile as a company. How Anthropic answers this question in their enterprise documentation over the next 30 days will tell you whether they see Managed Agents as a developer acquisition play or as a genuine enterprise platform. Those two paths lead to completely different outcomes in 2027. Three things to watch in the next 30 days. Pricing. Anthropic has not published pricing for Managed Agents yet. Usage-based pricing signals developer targeting. Platform fee plus usage signals enterprise targeting. Whichever they pick will reveal who they're actually selling to. Named reference customers. The first three enterprise reference customers Anthropic cites will tell you whether they have enterprise credibility for this move. Watch the Anthropic blog through mid-May. OpenAI's response. OpenAI will ship something comparable within 60 days. They have to. How fast they respond - and whether it's a feature match or a genuine platform strategy - will tell you how seriously they are taking this. Yesterday Anthropic was a model company with a platform ambition. Today they are a platform company with a model at the center. The difference matters more than most of the coverage this week will capture. About the author: Arpy Dragffy is founder of PH1 Research and co-host of the Product Impact Podcast. All claims about competitive positioning in this piece are based on public product documentation from the companies referenced. How helpful was this article? Founder, PH1 Research · Co-host, Product Impact Podcast Hosted by Arpy Dragffy and Brittany Hobbs. Arpy runs PH1 Research, a product adoption research firm, and leads AI Value Acceleration, enterprise AI consulting. Get AI product impact news weekly

Create a DSPy pipeline. 03 Apr 2026 READ TIME: 1 MIN Orchestrating AI-Driven DevOps Pipelines with Enhanced Compliance Automation Introduction In today's fast-paced digital landscape, organizations are under immense pressure to deliver high-quality software products quickly and efficiently while ensuring compliance with regulatory requirements. DevOps pipelines have become a crucial component in achieving this goal, but traditional approaches often fall short in providing the level of automation and compliance required. This is where AI-powered agents come into play, revolutionizing the way B4n1 approach DevOps pipeline orchestration and compliance automation. The Challenge of Traditional DevOps Pipelines Traditional DevOps pipelines rely on manual processes and scripting to automate tasks, which can lead to: * Inefficiency: Manual processes are prone to errors and can be time-consuming. * Inconsistency: Scripts can become outdated and may not account for changing infrastructure or compliance requirements. * Lack of Visibility: It can be difficult to track the status of pipeline execution and identify bottlenecks. Introducing Kiro AI-Powered Agents Kiro AI-powered agents are designed to address these challenges by providing a more efficient, consistent, and transparent approach to DevOps pipeline orchestration. These agents use machine learning algorithms to analyze infrastructure and compliance data, enabling them to make informed decisions and automate tasks. Integrating Kiro Agents with DSPy and LangChain To create an autonomous infrastructure vision and continuous compliance automation framework, B4n1 will integrate Kiro AI-powered agents with DSPy and LangChain. * DSPy: DSPy is a Python library for building and managing data pipelines. It provides a simple and efficient way to process and transform data, making it an ideal choice for integrating with Kiro agents. * LangChain: LangChain is a Python library for building conversational AI applications. It provides a range of tools and APIs for natural language processing, making it an excellent choice for integrating with Kiro agents. Practical Code Example Here is a simple code example that demonstrates how to integrate Kiro agents with DSPy and LangChain: import dspy from langchain import LLMChain from kiro import KiroAgent pipeline = dspy.Pipeline # Create a Kiro agent agent = KiroAgent # Define a LangChain LLMChain llm_chain = LLMChain( model_name="code-davinci-002", max_length=2048, temperature=0.7,) # Define a DSPy task task = dspy.Task( name="example_task", description="Example task", pipeline=pipeline,) # Define a Kiro agent task agent_task = agent.create_task( name="example_agent_task", description="Example agent task", pipeline=pipeline,) # Define a LangChain task lang_chain_task = llm_chain.create_task( name="example_lang_chain_task", description="Example LangChain task", pipeline=pipeline,) # Create a pipeline with all tasks pipeline.add_task(task) pipeline.add_task(agent_task) pipeline.add_task(lang_chain_task) # Run the pipeline pipeline.run This code example demonstrates how to create a DSPy pipeline, a Kiro agent, and a LangChain LLMChain. It then defines a DSPy task, a Kiro agent task, and a LangChain task, and adds them to the pipeline. Finally, it runs the pipeline. Benefits of Integration The integration of Kiro AI-powered agents with DSPy and LangChain provides several benefits, including: * Autonomous Infrastructure Vision: Kiro agents can analyze infrastructure data and make informed decisions about pipeline execution, reducing the need for manual intervention. * Continuous Compliance Automation: Kiro agents can analyze compliance data and automate tasks to ensure continuous compliance with regulatory requirements. * Improved Efficiency: The integration of Kiro agents with DSPy and LangChain enables more efficient pipeline execution and reduces the need for manual scripting. * Increased Visibility: The integration provides real-time visibility into pipeline execution and identifies bottlenecks, enabling organizations to make data-driven decisions. Top 5-10 Most Used Commands/Snippets Here are the top 5-10 most used commands/snippets for this technology: * dspy.Pipeline: Creates a new DSPy pipeline. * kiro.KiroAgent: Creates a new Kiro agent. * langchain.LLMChain: Creates a new LangChain LLMChain. * dspy.Task: Creates a new DSPy task. * kiro.KiroAgent.create_task: Creates a new Kiro agent task. * langchain.LLMChain.create_task: Creates a new LangChain task. * pipeline.add_task: Adds a task to the pipeline. * pipeline.run: Runs the pipeline. * kiro.KiroAgent.analyze: Analyzes infrastructure data. * langchain.LLMChain.generate: Generates text based on input. Visualization Here is a simple visualization of the pipeline using Chart.js: { "type": "bar", "data": {"labels": ["DSPy", "Kiro Agent", "LangChain"], "datasets": [ { "label": "Pipeline Execution Time", "data": [10, 20, 30], "backgroundColor": ["#FF6384", "#36A2EB", "#FFCE56"]}]}, "options": {"title": { "display": true, "text": "Pipeline Execution Time"}, "scales": {"yAxes": [ { "ticks": { "beginAtZero": true}}]}}} This visualization demonstrates the execution time of the pipeline for each task. Conclusion In conclusion, the integration of Kiro AI-powered agents with DSPy and LangChain provides a powerful framework for autonomous infrastructure vision and continuous compliance automation. By leveraging the strengths of each technology, organizations can create more efficient, consistent, and transparent DevOps pipelines that ensure continuous compliance with regulatory requirements. Have questions? Let's talk about how to apply these technical concepts to your workflow.