Agent Conf 2026
as the organisers put it
Join 500+ engineers in Warsaw Sept 17-18 2026 for Agent Conf. Learn agentic development, orchestration, and AI-driven workflows. Hosted by Callstack.
| title | author | length |
|---|---|---|
| Agentic Monitoring in Production | Sujith N S, Leonel Candido da Silva | 21m |
eBay deploys two production agents — a diagnosis agent that turns docs and runbooks into a structured investigation plan and an enrichment agent that queries… | ||
| AI Investigations Are a Distributed System Problem | Christopher Burns | 16m |
System pins scans pin to a stable commit then runs hundreds of deterministic rules to produce a reproducible hit list, builds a plan, and dispatches bounded… | ||
| Future of Software Development | Chris Trag, Margaret Pieczkowski, Grzegorz Wołodko, Bartek Pucek, Carl Ross | 51m |
Panel claims agents now write and review most code, turning human PR review into a cost/scale problem and pushing teams to agent-led PR review. Tools mentioned… | ||
| Beyond Components: The Future of AI UI | Ruben Casas | 21m |
Models change dramatically in late 2025 (CHP 5.2 and Opus 4.5) and move from copy‑paste snippets to generating complete apps from a single prompt. Three UI… | ||
| How Will We Prompt AGI? A History of Harness Hacks | Ivan Charapanau | 20m |
A concise history of harness hacks traces concrete wins and tradeoffs: Viktor runs a single continuous agent with 173 integrations, 5,400 tools, 1,300 skills… | ||
| Building a Neobank With Agentic Engineering | George Nikolaropoulos | 21m |
Natech reshapes product and engineering around agents and moves specs and implementation plans in the repository (GitHub issues and MD files) so regulatory… | ||
| The End of Apps | KITZE | 24m |
Argues that apps fragment tasks and predicts an agentic OS that runs 95% of work and nudges humans only for decisions. Describes building Benji, Android Tasker… | ||
| Agent Observability Beyond Vibes | Sven Großmann, Ivana Huckova | 21m |
Grafana Assistant serves more than 18,000 organizations and Grafana builds an observability stack that measures four trust dimensions: performance, behavior… | ||
| Goodbye Slop; Welcome Determinism | David Khourshid | 21m |
Argues that replacing mega‑prompts with explicit state machines removes agent “slop” by separating deterministic rules from generative work and makes behavior… | ||
| The Agent Control Plane: Managing Thousands of AI Agents | Nishant Gupta | 22m |
Operating fleets of tens to hundreds of thousands of agents requires a separate control plane that provides seven control functions (lifecycle, policy… | ||
| No Token Budget: Agentic Software Factories on Kubernetes | Carl Ross | 19m |
OpenAI runs an exploratory system where agents own the full lifecycle: agents write, review, and release code and run 24/7 for months without human… | ||
| The Critical Advantage With AI | Tejas Kumar | 22m |
Claims that the critical advantage is trainable skill — taste defined as tacitly judging quality against external standards — and that taste improves with… | ||
| Has This Happened Before? Queryable Memory for Database Logs | David Budáč | 22m |
Operational logs from a fleet of 300 databases move on‑premises into Elasticsearch ingestion where deterministic Python compaction and local models produce a… | ||
| Quantization impact on Apex performance | Artur Morys-Magiera, Lech Kalinowski | 20m |
Quantized Apex (Gemma 4 31B instruct) is evaluated on a DGX Spark across nine variants: 27 GPU hours, 4,344 runs, 6M input tokens, 891k outputs. BF16 is 61.4… | ||
| TTT, TAO, δ-mem: When to Spend a Gradient | Anjaneya Sastry Kappagantu | 23m |
Compares three concrete methods and when to pay for a gradient: Test-Time Training is ephemeral per inference, Test-Time Adaptive Optimization is offline… | ||
| The AI Agent Iceberg | Zuzanna Skoczylas | 20m |
Sierra runs production agents for 40% of the Fortune50 and shows concrete engineering: an intent classifier that unlocks context dynamically (progressively… | ||
| Agents FC: Harness Agents Into a Winning Team | Anisha Malde, Kourtney Meiss | 22m |
A concrete recipe for turning fragile multi‑agent systems into observable, governed teams: build a harness with runtime, observability and governance layers… | ||
| Agentic Coding at scale at Open Mercato | Piotr Karwatka | 21m |
Open Mercato runs an agentic software factory that produces about 1,200,000 lines of code, ships biweekly, and coordinates 140+ contributors. It replaces a… | ||
| Build an Agent with Eve: The Open Agent Framework | Dom Sipowicz | 22m |
Eve is an open-source agent framework from Vercel that builds durable, long-running agents with one-folder conventions: start with npx eve init and a single… | ||
| Your Biggest User Isn't Human: Designing DevTools for AI Agents | David Mokos | 19m |
Expo sees 50% of CLI commands executed by coding agents and builds a pipeline that turns agent friction into fixes: an agent-facing feedback surface yields… | ||
| Stop Making LLMs Deterministic: A Spec-Driven Software Factory | Jakub Kuzimski | 23m |
Allegro pilots a spec-driven factory that treats the spec as source of truth and generates code via a multi-agent pipeline (planning, Geneva coding agent… | ||
| Building a Software Factory with Cursor, xAI, and Grok Bot | Nick Miller | 27m |
Grok Bot is a general-purpose agent that runs on a cloud computer with a persistent browser, connects to Gmail, Slack, Notion, Salesforce and calendars, and… | ||
| Keynote | Mike Grabowski, Michał Pierzchała | 22m |
Callstack replaces hand‑coding with agent orchestration using a CLI library called Agent Device, runs parallel agents via git worktrees and cloud agents, and… | ||
| Designing for Growing Intelligence | Levan Kvirkvelia | 19m |
Rork moves from heavy guardrails to an architecture that benefits when models improve: Opus 4.5 eliminates many hand-coded checks and enables Swift code… | ||
| I Built Your Personal Software Ecosystem | Kent C. Dodds | 22m |
Kody converts prompt skills into deterministic packages (demo: a PR-ready package that any agent can call), exposes only two agent tools (search and execute)… | ||
| nothing matches | ||