AI-fokus 2026
| title | author | length |
|---|---|---|
| Agentic AI Coding Practices for Speed with Quality | Adam Tornhill | 45m |
Claims that production work is fully agentic with 100% production code written by agents while measuring roughly 2–3x faster task completion. Independent… | ||
| AI agents in practice - beyond the prototype | Henrik Kniberg | 45m |
Describes building agents that move from flashy prototype to production: a spreadsheet screener processes thousands of rows and improves decision quality while… | ||
| Spec-driven Development: How AI Changed Everything (And Nothing) | Simon Martinelli | 44m |
Argues for AI Unified Process and spec-driven development that keeps structured requirements as the source of truth and uses AI to generate diagrams, entity… | ||
| Taming Agentic AI: Run Rogue Code Safely on Kubernetes with Agent Sandbox | Abdel Sghiouar | 14m |
Kubernetes-native Agent Sandbox runs untrusted agent-generated code inside ephemeral pods backed by a CRD-based template system and a Python client SDK that… | ||
| Stop Stuffing Your LLM's Context Window | César Soto Valero | 15m |
Claim: bigger context windows are not a magic fix and real failures come from how information is fed. Models today support up to about 1,000,000 tokens (an AI… | ||
| Making AI Adoption Work: A Practical Structure for Modern Software Development | Cecilia Borg | 46m |
Practical structure shows that AI collapses coding time and shifts bottlenecks to testing, review, codify (versioned context), and planning; agents run… | ||
| Context Is the New Code | Patrick Denis | 28m |
Claim: treat context like code with a repeatable Context Development Lifecycle (Generate, Evaluate, Distribute, Observe). Demonstrated tools and patterns… | ||
| Zero Secrets: Building Secure Infrastructure for AI Agents | Jonathan Ström | 34m |
Compares three concrete architectures — naïve host-privilege, MCP tool control, and sandbox + control plane — and argues infrastructure-level containment is… | ||
| Own Your LLM: Running Generative and Agentic AI Locally | John Davies | 47m |
Run open-weight LLMs locally on phone, laptop or private server and keep data private while cutting latency to milliseconds and throughput to 200–300… | ||
| How Does a Coding Agent Work? | Tore Nestenius | 44m |
Explains a working coding agent built as a while loop that sends a system prompt (~25,000 chars), tool descriptions (about 90,000 characters of tool… | ||
| Fullstack AI: The Building Blocks for Production Ready AI-Powered Software | Sebastian Nilsson | 46m |
Claims a production-ready AI agent requires explicit context management, tool calls and memory rather than a bigger LLM; builds a demo research agent using… | ||
| AI fokus 2026 intro | Matias Carlsson, Helena Ylitalo | 02m |
Short stage intro that opens with a brief musical opening followed by a two person welcome and no technical claims or demos The segment contains no slide… | ||
| Agentic workflows that compound | Magnus Gille | 38m |
Argues that agentic workflows compound by combining three concrete practices: a two-layer persistent memory (repo status.md snapshot plus a self‑hosted Munin… | ||
| Beyond the Vibe, How far can we push It? | Roger Johansson | 42m |
Claim: a complete ECMAScript‑compliant JavaScript engine is built using AI agents and C# with ECMA-262 94,000 tests as the source of truth. Tools and pipelines… | ||
| Mythos & Glasswing, What we actually know | Dan Bergh Johnsson | 42m |
Anthropic’s Mythos demonstrates a genuine jump: it finds 271 vulnerabilities on Mozilla’s test bench, completed the long‑running cyber‑range challenge in ~17… | ||
| The Terminal Strikes Back, Agentic Engineering using the CLI | Andreas Wänqvist | 48m |
Demonstrates GitHub Copilot CLI as a terminal-first agentic dev environment with background delegation to GitHub cloud, session resume and checkpoints, and a… | ||
| Beyond context limits: why a picture is worth a thousand words | Oleksandr Kachur | 13m |
Argues that embedding images as optical context yields far fewer tokens and faster inference than raw text for observability data, showing a Grafana demo where… | ||
| What Happens After the POC: Building Agentic Systems That Last | Markus Mäkelä, Deniz Kuznetsov | 41m |
InfraAssistant is a Kubernetes microservice hooked to Slack that starts as a simple RAG over internal docs and scales into a dynamic multi‑agent system using… | ||
| AI Testing Beyond the Basics: Ensuring Truthful and Reliable Agents | Mario-Leander Reimer | 46m |
Argues that unit tests fail for probabilistic agents and demonstrates a runnable testbench built on Testkube + Ragas (with DeepEval style verification)… | ||
| A Love Letter to Code – What Happens When We Stop Writing Most Of It Ourselves? | Henrik Ygge | 34m |
Argues that AI doesn’t replace developers but amplifies existing weaknesses: lack of explicit rules, messy boundaries, and undocumented knowledge. Shows… | ||
| Idea to deployed agentic system with Akka Specify | Patrik Nordwall | 15m |
Claim: a plain-English requirement becomes a deployed agentic service without hand-writing code. Workflow uses Akka Specify inside Claude Code plus the Akka… | ||
| How to be a great leader for your Agentic teams | Björn Wismén | 15m |
Company runs its entire org as AI agents and declines a client ISO certification integration after a 45 second board debate, proving decisions can be faster… | ||
| From Prompt to Multi Agent System: The evolution of our AI product | Hila Fox | 40m |
Qodo describes moving from a single IDE coding agent to an agentic system and now a multi-agent code review that runs in a Python backend and serves enterprise… | ||
| From Pilots to Practice: Leading the agentic AI transformation | Juho Nevalainen | 43m |
Argues that the real bottleneck is planning and that treating Plan→Build→Run as one continuous engine shortens time‑to‑value, backed by a simple diagnostic and… | ||
| Your Application as an MCP Client | Cecilia Wirén | 41m |
A C# console MCP client is built that connects to a local MCP server, fetches a dynamic list of tools and prompts, sends those tool descriptions into the chat… | ||
| nothing matches | ||