Recommended Free Tools
QCon San Francisco 2026’s program examines a practical shift in production engineering: AI agents are starting to act as users of customer systems, observability tools, and software delivery workflows. The central question is not simply what an agent can do, but what authority it should have, what evidence should precede its actions, and where engineers must retain control.
InfoQ reports that the conference runs November 16–20, 2026, at the Hyatt Regency San Francisco, with conference sessions November 16–18 and training November 19–20. The program examples offer a useful lens on agent safeguards, coding-agent verification, machine-readable observability, and the operational trade-offs behind distributed systems.
What QCon San Francisco 2026 covers
The event’s distributed-systems program connects latency, consistency, observability, capacity, and failure handling to the everyday trade-offs of operating systems at scale. Its agent-related sessions bring that same production focus to new kinds of system users: agents that can query operational data, generate code, or initiate actions affecting customers.
InfoQ’s October 2, 2026 coverage is a selective program overview rather than a complete session catalog. The session examples below are best read as engineering questions and case studies, not as a claim that every relevant approach or talk is represented. InfoQ’s QCon San Francisco 2026 program coverage provides the reported details.
#1 Best Overall
- High-Performance AI Processor：The MS-02 Ultra features an Intel Core Ultra 9 285HX (24C/24T, up to 5.5 GHz, 13 TOPS NPU), delivering fast and efficient performance for AI inference, algorithm development, and media workloads. A PCIe x16 expansion slot supports desktop-class GPU upgrades for advanced model training and accelerated computing tasks. It's ideal for creators, engineers, and teams handling intensive parallel workloads.
- 4 × M.2 PCIe 4.0 + 4 × DDR5 SODIMM slots:Four DDR5 SODIMM slots support up to 256 GB of memory, while ECC helps maintain data integrity in mission-critical environments. Four PCIe 4.0 M.2 slots support up to 24 TB of storage, supporting RAID 0/1/5/10, combining high-speed performance with data protection. It allows for the creation of independent scratch disks, media libraries, and project drives, providing high-throughput for production workflows.
- PCIe & USB 4.0 v2: Up to three PCIe slots can be equipped, including a dual-slot x16 GPU. The main slot supports PCIe 5.0, meeting the needs of high-bandwidth creative and computing workloads. USB 4.0 v2 (80Gbps) supports high-bandwidth external storage and displays.
- Ultra-fast Networking: Wi-Fi 7 further enhances wireless performance with next-generation speeds and low-latency stability. Intelligent bandwidth switching optimizes throughput in different network environments, ensuring optimal performance for enterprise or local networks. Dual 25GbE ports (providing up to approximately 3.125 GB/s bandwidth, about 25 times faster than traditional 1GbE), enabling seamless large-scale file transfers and parallel computing. 10GbE and 2.5GbE ports, with support for Intel vPro technology, ensure enterprise-grade remote management and deployment flexibility.
- Server-grade thermal architecture: Utilizing a dedicated CPU/GPU airflow design, equipped with a 6-pipe dual-fan cooler, it maintains stable performance even under sustained loads, delivering up to 140W Turbo power while maintaining a 100W TDP, and operating with noise levels as low as 36 dB. An integrated 350W power supply ensures stable and reliable output for demanding computing tasks and fully loaded extended configurations.
How much authority should a customer-facing agent have?
Airbnb’s layered safeguards
In “How Airbnb Guardrailed Its AI Customer Support Agent,” Airbnb Distinguished Engineer Weiping Peng discusses an agent described by InfoQ as serving millions of customers, preserving context across conversations, and initiating account actions. InfoQ reports safeguards including input sanitization, classifiers, shadow testing, false-positive management, and rapid-response mitigations. These are figures and descriptions attributed to InfoQ’s October 2, 2026 article, not independently audited measures.
The engineering implication is that authority should be proportional to consequence. Drafting a response for an employee to review is materially different from changing an account. A control system for consequential actions needs layers: reduce harmful inputs, detect risky requests, test behavior before exposure, monitor errors such as false positives, and have a way to mitigate failures quickly once the system is live.
Shadow testing is especially useful as a bridge between offline evaluation and customer impact: the system’s behavior can be assessed without allowing its proposed actions to affect the live workflow. The program description does not specify Airbnb’s exact thresholds, classifier design, or action-approval policy, so those should not be inferred from the session summary.
What evidence should agent-generated code need before release?
OpenAI’s coding-agent case study
“Lessons from Building a $100M Product in Six Weeks at OpenAI,” by OpenAI Member of Technical Staff Brian Yang, concerns building OpenAI Ads with coding agents. InfoQ reports the result as more than $100 million in annual recurring revenue in under six weeks; this is the article’s account of the case study, not an independently validated performance claim. The session is described as covering feedback loops, verification, token economics, and decisions that remained under human ownership.
Do these 3 things before closing this tab:
1Repair Windows errors before they cause bigger problems2Fix the driver behind crashes, sound loss and screen glitches3Clear out junk files and repair common Windows errorsThe production lesson is to separate implementation speed from release confidence. Agents may increase the rate at which code is proposed, but that does not establish that a change is correct, secure, maintainable, or appropriate to ship. Teams still need evidence suited to the risk of the change: review, tests, operational checks, and a human decision about architecture and release. The session description does not specify a universal verification process, and no single checklist should be assumed to fit every system.
Human ownership matters most where choices have durable consequences: system boundaries, quality standards, production risk acceptance, and whether a change should be released. Delegating code production is not the same as delegating accountability.
Rank #3
- Professional AI & Creator Workstation: AMD Radeon AI PRO R9700 GPU with 32GB GDDR6 is engineered for AI development, professional content creation, and compute-intensive workloads.
- Massive 32GB Memory Capacity: 32GB of GDDR6 memory on a 256-bit bus provides ample bandwidth for large AI models, 8K video editing, and complex 3D rendering.
- Advanced RDNA 4 with AI Accelerators: 64 Compute Units with 3rd Gen Ray Tracing and dedicated 2nd Gen AI Accelerators for groundbreaking AI performance and visual computing.
- Professional Blower Cooling: Efficient single blower design exhausts heat directly out of the chassis, ideal for multi-GPU workstation and server configurations.
- Enterprise-Grade Thermal Solution: Vapor chamber heatsink with industrial Honeywell PTM7950 thermal interface material ensures reliable cooling under sustained professional loads.
How can production context be made usable to agents?
Honeycomb’s observability MCP
In “Making Production Legible to Agents: Lessons From Building an Observability MCP,” Honeycomb Technical Fellow Liz Fong-Jones discusses an MCP server for running production queries through agents. InfoQ reports that more than 40% of Honeycomb’s weekly active users use it for those queries. The figure is attributed to InfoQ’s October 2, 2026 coverage and is not independently audited.
The reported topics include token economy, tool descriptions, schemas, evaluations, output formats, and defects discovered through real-world use. Together, these point to a broader design requirement: giving an agent access to data is not enough. The interface must make the available operations and their results understandable, constrain what can be queried or changed, and be evaluated against actual use rather than judged only by whether a tool call succeeds.
For production systems, legibility is also a safety concern. If an agent receives incomplete or confusing context, it may draw an unsound conclusion even when the underlying telemetry is correct. Tool schemas, descriptions, and output formats therefore shape both usefulness and risk. The session summary does not establish the MCP’s full permission model or the precise evaluation methodology.
Rank #4
- FAST RUNS IN THE FAMILY — The 16-inch MacBook Pro with the M5 Pro or M5 Max chip brings next-generation speed and powerful on-device AI to personal, professional, and creative tasks. With all-day battery life, double the starting storage,* and a breathtaking Liquid Retina XDR display, it’s pro in every way.*
- BUCKLE UP — Along with a next-generation CPU, faster unified memory, and up to 2x faster SSD storage,* M5 Pro and M5 Max feature a more powerful GPU with a Neural Accelerator built into each core, delivering faster AI performance and on-device training capabilities. So you can blaze through demanding workloads at mind-bending speeds.
- BUILT FOR AI — Apple silicon, and every major component that powers it, is designed to run demanding on-device AI workloads like LLM inference and training. And Apple Intelligence helps you write, express yourself, and get things done effortlessly with groundbreaking privacy protections at every step.*
- ALL-DAY BATTERY LIFE — MacBook Pro delivers the same exceptional performance whether it’s running on battery or plugged in.*
- MACOS RUNS APPS FAST — All your go-to apps run lightning fast in macOS, including built-in apps like FaceTime and Messages. Plus, built-in virus protection and free software updates help keep your Mac running smoothly and securely.
Why infrastructure optimizations need system-level evaluation
Netflix’s adaptive compression trade-offs
“Orderly Keys, Wild Values: Adaptive Compression for Distributed Key-Value Storage,” by Netflix engineers Joseph Lynch and Ayushi Singh, addresses compression in a system described by InfoQ as handling billions of daily requests and petabytes of key-value data. Those scale figures are reported context, not independent measurements.
The session’s reported trade-offs include storage footprint, cache behavior, network I/O, p99 latency, dictionary versioning, compatibility, and rollout safety. That is a reminder that reducing stored bytes is not automatically a system-wide win: compression can affect how data fits in caches, how much work is needed on the request path, and whether different versions of components can interpret data safely during deployment.
Teams evaluating a similar change should define success across the coupled dimensions that matter to their workload, not just storage use. They also need to account for compatibility and staged rollout, because a locally beneficial format change can become an operational risk if producers, consumers, or stored data move out of sync. The program summary names these considerations but does not provide benchmark results or a recommended compression design.
Windows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallOutdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchBest Value
- 【High-Performance APU】The MS-S1 MAX features an AMD Ryzen AI Max+ 395 APU, integrating a Zen 5 architecture CPU (up to 5.1GHz, 16C/32T, 64M L3 Cache), an RDNA 3.5 GPU, and an NPU (50 TOPS). The total system output is 126 TOPS. It provides powerful parallel computing capabilities for demanding AI workflows. It is ideal for running local LLMs, multimodal models, and computationally intensive tasks
- 【128GB UMA Memory】Equipped with up to 128GB of LPDDR5x-8000MT/s unified memory, it enables the CPU and GPU to access a shared, high-bandwidth memory pool with extremely low latency. Ideal for large-scale AI inference, 3D workloads, and complex timelines in video editing. It eliminates traditional VRAM bottlenecks, ensuring smoother data transfer during high-intensity computations. The UMA design maximizes performance stability under high loads
- 【Flexible Expansion】The MS-S1 MAX features USB4 V2 (up to 80Gbps), dual 10GbE LAN, HDMI 2.1 (up to 8K60), a full-length PCIe x16 expansion slot, and dual M.2 slots supporting up to 16TB RAID 0/1. Wi-Fi 7 provides stronger signal coverage and a more stable wireless experience. The slide-out design facilitates upgrades and maintenance. It easily adapts to personal, studio, or rack-mount enterprise environments
- 【High-Efficiency Cooling System】Utilizing an aerospace-grade aluminum alloy chassis, copper base plate, six heat pipes, dual turbine fans, and advanced PCM thermal conductive material, it maintains stable cooling performance even under continuous load. This system supports 130W continuous power and 160W peak power operation, with a built-in 320W power supply. It boasts multiple global certifications including CCC, FCC, UL, CE, and UKCA, ensuring stable and reliable operation in various environments
- 【Cluster Design】Two MS-S1 MAX units can be configured as a dual-unit cluster to run a large 235B Q4 model locally, achieving an output speed of 10.87 tok/s. Supporting 2U rack deployment, multiple MS-S1 MAX units can be cascaded into a distributed cluster to create a high-efficiency AI computing center. A cluster of four MS-S1 MAX units successfully ran a DeepSeek-R1 671B Q4 large model. A reserved cluster power-on interface allows for unified start-up and shutdown
What the program suggests engineers should retain
Across these examples, the boundary between delegated work and human responsibility depends on the consequence of an action and the evidence available to assess it. Customer account changes call for safeguards beyond those needed to draft text; generated code still needs verification and human release ownership; and agent-facing operational tools need clear constraints and evaluation.
- Match an agent’s authority to the impact of the action it can take.
- Use layered controls, including prevention, evaluation, detection, and rapid mitigation, for customer-facing actions.
- Do not treat faster code generation as evidence that architecture, quality, or release decisions can be delegated without oversight.
- Design operational tools so agents can interpret context and constraints, then test those tools against real usage and defects.
- Evaluate infrastructure changes across latency, capacity, compatibility, observability, storage, and rollout risk rather than a single local metric.
Event dates, training, and registration details
InfoQ reports QCon San Francisco 2026 at the Hyatt Regency San Francisco from November 16–20. Its coverage separates the main conference, November 16–18, from training on November 19–20. InfoQ’s conference listing also lists QCon San Francisco for November 16–20, 2026.
The October 2 coverage says early-bird conference tickets are $2,955 through October 13, 2026. It also describes a four-day InfoQ Certified Architect Program, including a peer cohort and a half-day workshop on November 19, plus optional hands-on training on November 19–20. These are details reported in that article, and registration prices, deadlines, schedule, and availability can change; check the current event information before making plans. The coverage does not provide a separate ticketing-page link or establish the current status of these offerings.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
