Articles by Jon
-
The December Dead Zone
The December Dead Zone
One of the many things sales people do on a daily basis is to manage their calendars for optimal sales results. Aside…
7 Comments
Activity
11K followers
-
Jon Sanchez reposted thisAgents are dynamic and by nature probabilistic. Thats the power they bring in but you also need determinism and explainability so that you can trust them for your needs (especially if you are building something for your work). And you need to be sure that your agents are running successfully (or failing gracefully) vs randomly crashing. That means you need durability in its execution. And that IS the difference between an agent that runs great on your laptop during a demo and a successful production deployment of it Viren Baraiya will be talking about this in-depth in the context of building an Agent Harness at the AI Engineer World Fair next week in San Francisco. If you are there, don't miss it. We will have more folks from our team attending as well. Please feel free to reach out to Deepti Reddy Nicholas Cole Nikhil Kala Adrian Warman Richie Niu Rohan K. T Yiting Ling if you want to chat moreJon Sanchez reposted thisDynamic Workflows Need a Durable Harness AI agents increasingly generate and revise workflows at runtime. But how do you safely execute and recover a workflow that did not exist until the agent created it? At AI Engineer's World Fair at San Francisco next week (July 1, Expo Stage 1, 11:10am), Orkes CTO Viren Baraiya will show how Orkes Conductor provides a durable execution harness for agent-generated workflows with retries, parallelism, human approvals, MCP tool calls, policy enforcement, and safe side effects using idempotency and saga compensation. The agent owns the plan. The harness owns the guarantees. Hope to see you there! https://lnkd.in/gRXhnpRR
-
Jon Sanchez shared thisPlease join us in SF for a happy hour during the AI Engineer World Fair. July 1, 2026 starting at 530. See link for details and register. https://luma.com/3ftf8emgHarness the Agent. Happy Hour by Orkes @ Chotto Matte · LumaHarness the Agent. Happy Hour by Orkes @ Chotto Matte · Luma
-
Jon Sanchez shared thisWhat happens to your AI agent when the process it runs in crashes? Come to the live session and fine out.Jon Sanchez shared thisWhat happens to your AI agent when the process it runs in crashes? If the answer is "it dies and we start over," you're not alone — and you're not in production-ready territory. Join us this Thursday for a live session on durable AI agents. We'll kill agents mid-execution on stage, resume them from the exact step, and show how Agentspan (open sourced by Orkes) adds crash recovery, retries, HITL pauses, and full observability to your existing agents. With Agentspan, you can build, run, and observe agents with durability as a first principle from day one. Or bolt it onto existing frameworks like LangGraph, OpenAI Agents SDK, or Google ADK in a single line of code. https://lnkd.in/gPKcgxt9 Two sessions to cover every region: 11 AM CT for Americas/EMEA, 7 PM CT for APAC. See you there!Webinar: Beyond Sandboxes: Architecting Durable Runtimes for AI AgentsWebinar: Beyond Sandboxes: Architecting Durable Runtimes for AI Agents
-
Jon Sanchez shared thisLooking forward to seeing you there.Jon Sanchez shared thisAction-packed week ahead: 3 events, 3 talks, and we're kicking things off tomorrow. Join us in Boston at QCon Software Development Conferences for QCon AI, where our co-founder and CTO Viren Baraiya will be speaking on a topic that hits at the heart of production AI infrastructure: "Beyond Sandboxes: Architecting Durable Runtimes for AI Agents" Most agent infrastructure focuses on sandboxing, keeping tool execution safe. But what protects the agent itself? When the agent loop lives in memory, a crash takes everything with it: state, context, explainability. Viren will walk through what a durable runtime actually looks like: replay, traceability, crash recovery, and operational clarity, using Conductor and Agentspan as real-world examples. 📅 Monday, June 1 | 3:40 PM EDT | East Balcony 👉 https://lnkd.in/gVBe47r9 See you there. More events to announce later this week. 👀QCon AI Boston 2026 | Beyond Sandboxes: Architecting Durable Runtimes for AI AgentsQCon AI Boston 2026 | Beyond Sandboxes: Architecting Durable Runtimes for AI Agents
-
Jon Sanchez reposted thisJon Sanchez reposted this📢 You do not need another framework or SDK for agents. What you need is a runtime platform capable of running your agents and explaining their decisions a year later when required. This is precisely what we are achieving with Agentspan, which harnesses the power of Conductor OSS. Durability is essential, but can your agent runtime platform explain the decisions made by the agents? Or assist you in doing so. https://lnkd.in/gD_WWt-X
-
Jon Sanchez reposted thisJon Sanchez reposted thisWe build guardrails as a first class primitive in Agentspan with Conductor. Backed by Conductor as core orchestration engine. When you set the guardrail and make it programmatic - it will execute and if not allowed, the workflow terminates rather than sending it back to LLM to judge and decide. Give it a spin https://lnkd.in/gD_WWt-X
-
Jon Sanchez liked thisJon Sanchez liked thisAd astra ⭐ At 7:26am ET (1126 UTC) on Aug. 30, the Nancy Grace Roman Space Telescope took to the skies en route to its new home 1 million miles away from Earth. It should reach its final destination in one month: https://lnkd.in/euaZVKBX
-
Jon Sanchez liked thisJon Sanchez liked thisWhy do I eat 3 eggs and salmon for breakfast? Here’s why: → Because I don’t choose my breakfast just based on how it makes me feel right after I eat. I care about how hungry I’ll be 2–3–4 hours later. Because the hungrier I get, the more time and energy I waste thinking about food. And if I don’t wanna feel hungry again after 2 hours, I need to make sure my breakfast is rich in protein. Why? Because protein is the most satiating macronutrient. Research shows that when we eat a high-protein breakfast, we can: 1. Feel full for longer 2. Reduce hunger and cravings throughout the day Why is this? Because when we eat protein, our digestive system doesn’t just process the food. It also stimulates several signals involved in appetite, like GLP-1, for example. These signals tell your brain → “Hey, food has arrived! You can start feeling full” But here's the thing: these signals are almost invisible when you don’t eat protein. So if you eat something like cereal and orange juice, your brain still receives fullness signals. But compared with a protein-rich meal, those signals are weaker, and you’re gonna feel hungry again after 2 hours. Think of it like turning up the volume. Protein turns up some of the signals that tell your brain → “We’ve eaten. We’re good.” That’s why a protein-rich breakfast keeps you fuller for longer. And that’s why I eat 3 eggs and 200 g of salmon for breakfast (with veggies, avocado, olive oil, and a bit of quinoa). It gives me roughly 60 g of protein. And there’s no way I feel hungry for 4 hours. Which basically means no cravings. And no cravings mean I can focus on getting things done instead of thinking about food all day. Remember: if you want stable energy at work (and lose weight), your number 1 goal is learning how to control your hunger. Eat a high-protein breakfast (with no sugar), and you won’t feel hungry for 3–4 hours.
-
Jon Sanchez liked thisJon Sanchez liked this6 weeks after giving birth, I couldn’t wait to start working again. Not because I had to. And definitely not because I don’t love spending time with my daughter. But because I genuinely wanted to. For a moment, it almost felt like I should feel guilty for admitting that. Then I realized something: For years, one of my biggest goals wasn’t to work less. But it wasn’t to work more either. It was to build work I’d actually miss when I stepped away from it. Work that challenges me. Work that gives me energy. Work that feels meaningful to me. And stepping away from it made me realize: I had actually achieved one of the things I wanted most. I had built work I genuinely wanted to come back to. I don’t think the goal is to love every minute or every day of your work. I certainly don’t. But there’s something pretty special about building work you genuinely want to be part of your life. I’m still figuring out how I want work and motherhood to fit together in this new chapter. Maybe I’ll share more about that journey here. At the same time, I’m very aware of what makes this choice possible for me: → A partner who takes on a huge share of the care work → A lot of freedom in how I structure my work and my time But one thing I already know: Loving my work doesn’t compete with loving my daughter. It simply means they’re both parts of my life that matter deeply to me 🤍 To all the self-employed mothers here: What has helped you find a setup that works for you?
-
Jon Sanchez liked thisJon Sanchez liked thisWe pointed Crusoe Managed Inference at Wiz's Day-One CTF. With the right browser tooling and a run-code primitive, the agent solved all seven challenges in less than 2 hours for ~$13, showing a strong harness matters more than a frontier model.Solving the Wiz Day-One CTF with Crusoe Managed InferenceSolving the Wiz Day-One CTF with Crusoe Managed Inference
-
Jon Sanchez liked thisJon Sanchez liked thisAfter 12 years with Walmart, I'm starting a new chapter in my career… A group of us unexpectedly became associates of Walmart in 2014. Our former employer went bankrupt on a Wednesday in March; Walmart, interested in bidding for the software assets, flew a team to Reston, Virginia, to meet with us the following Monday. As part of its bid for the assets, Walmart offered us employment. At the time it was a bit of a leap of faith — I didn't know much about the company beyond what the team from Walmart conveyed to us about its culture and values. I had some reservations. My boss at the time told me, "You never know how these things will turn out," and he was absolutely right — I ended up being part of a great company for 12 years. In my tenure at Walmart, in software engineering and program management leadership roles, I met and worked with incredible people, both talented and dedicated. My time at Walmart has been an incredible journey in both learning and personal growth. For my next chapter, I'm looking to continue my career in Program Management leadership, or a related field, based in the Northern Virginia area or open to remote work. During this transition, I'm using the opportunity to build my skills in Gen AI, pursue cloud certifications, and expand my knowledge more broadly. If you know of a role you think I would be a good fit for, or a company you really enjoy working at, please reach out. Thanks for being a part of my journey! Best regards, Joe #Walmart #OpenToWork #NorthernVirginia #TechLeadership
-
Jon Sanchez liked thisJon Sanchez liked thisI joined Span to lead sales and it has been a blast! I have spent the last several weeks meeting with CTOs and Engineering leaders navigating the shift from human driven development toward fully agentic software factories. Code throughput is exploding, token spend is climbing, and engineering leaders are asking a new set of questions about how they measure, understand, and optimize what's happening across their SDLC I could not be more excited about the role Span is playing at the center of this transformation! We're also coming off a record breaking quarter with a mighty sales team of five: Payton Andrews Bry B. Kevin Woods Matthew Lim John Popovich Stephen Poletto And we're hiring! If you're an AE or BDR and you're excited about the future of software development, I'd love to meet you. I wrote a little more about what I've learned in my first two months 👇
-
Jon Sanchez liked thisJon Sanchez liked thisProud to share this one. Six months ago I started a Substack. In July, I resigned from the corporate job I loved to go all in on my writing. When I started, I had no audience. No guarantee. Just a thing I had always wanted to do and a decision to finally stop waiting. Last Friday, I woke up to #87 rising writer in Health & Wellness on Substack. Hard work pays off. It doesn’t always happen fast. It doesn’t always look the way you thought it would. But it pays off. Keep going.
-
Jon Sanchez liked thisJon Sanchez liked thisIn a Toyota plant, a defective part physically cannot move to the next station. Sensors, jigs, shapes that only fit one way. They call it 𝒑𝒐𝒌𝒂-𝒚𝒐𝒌𝒆: mistake-proofing. AI agents need this more than any human team ever did. Agents generate volume, and volume includes hallucinated APIs, shallow tests, and confident nonsense. The fix is not better prompts. It's 𝐠𝐚𝐭𝐞𝐬 the code cannot skip: The patch doesn't reach the testing agent unless it compiles. It doesn't reach review unless lint and the security scan pass. It doesn't merge unless tests run green in a clean environment. None of these ask whether the agent seems right. They make wrong physically unable to travel down the line. If your agent output goes straight to human review, humans are your poka-yoke. That's the most expensive sensor there is. What's the most significant gate in your pipeline? #SoftwareFactory #AISDLC
-
Jon Sanchez liked thisJon Sanchez liked thisInstead of watching 2 hours of Netflix tonight, watch this Stanford lecture. It’s the clearest end-to-end explanation I’ve seen of how ChatGPT and Claude AI are actually built. From tokenization and BPE all the way to the transformer architecture, training pipeline, and next-token decoding. Whether you’ve never written a line of AI code or you’ve been shipping agents every day, this 2h 34min session will click things into place that most people take years to figure out. Together with this guide, you will be able to Build an Agentic OS with Claude Fable 5 in a few hours 🤖: https://lnkd.in/dZsST5qR Bookmark it and clear the time this weekend (link: https://lnkd.in/dG8d6jKr), because this might be the single most valuable thing you learn all month.
Experience & Education
Organizations
-
Delta Sigma Pi - International Business Fraternity
Officer
Recommendations received
33 people have recommended Jon
Join now to viewOther similar profiles
Explore more posts
Explore top content on LinkedIn
Find curated posts and insights for relevant topics all in one place.
View top content