Industry experts John Crickett and Gregor Ojstersek join us to recap their recent Ready, Spec, Ship hackathon where developers built projects using Kiro and spec-driven development. They'll walk through the winning entries, highlight what stood out, and share what they learned about how developers are approaching agentic workflows. Tune in live for Q&A.
Kiro
Technology, Information and Internet
Agentic AI development from prototype to production.
About us
Kiro helps you do your best work by bringing structure to AI coding with spec-driven development.
- Website
-
https://kiro.dev
External link for Kiro
- Industry
- Technology, Information and Internet
- Company size
- 10,001+ employees
Updates
-
A free year of Kiro, now for students around the world 🌎 With a new school year starting, we want more students heading back to campus to have serious tools in their hands. We are expanding the Kiro Students program to 121 new universities across 18 countries. Eligible students in those 18 countries now get the same offer: one full year of free access to Kiro with 1,000 credits per month, with full access to paid features such as premium models and Kiro Web. Sign up and learn more here 👉 http://spr.ly/6046BGypJG
-
Agentic engineering in the cloud ☁️ Kiro Web and cloud sessions are now generally available with cloud configuration. More of your work can keep running in the cloud while you move between the browser, CLI, IDE, and devices, with your configuration there too. Teams can scope sandbox access with IAM roles and authenticate through Okta or Microsoft Entra ID. Included with Pro, Pro+, Pro Max, and Power. Uses your existing Kiro credits, with no separate charge for cloud compute. Full breakdown 👉 http://spr.ly/6046BG6kly
-
-
Cloud Sessions let you move your development off your local machine, opening up longer tasks, bigger projects, and true unattended workflows. Join AWS Hero Johannes Koch as he walks through how cloud sessions feel in practice, where they shine, and where they still have rough edges. We'll build a small feature live and demo the full end-to-end workflow. Bring your questions.
Building beyond your laptop with Cloud Sessions
www.linkedin.com
-
How do you screen 27 prompt changes in a week and still know which result belongs to which change? I turned the Kiro team's detailed write-up into this one-page diagram. They write the candidate changes and use the results to decide what ships. An automated system runs the evaluation. The LLM judge scores internal developer conversations across 15 behavioral dimensions using explicit evidence, and then eligible developers stay in stable cohorts so each candidate can be compared with the control. The same rubric scores every group, and a combined cohort checks interactions. Small cohorts help locate regressions, but the team needs larger follow-ups before drawing broader conclusions. When the model changes, the team runs the checks again. That's why the 32% and 4% results are separate: they came from different model-and-prompt baselines. Full write-up from the eng + research teams: https://lnkd.in/gFh2iZua 👩🏻💻 Follow me ( Brooke Jamieson ) for the latest AI + AWS updates 📍 Save + Share 🏷️ AWS Developers #Kiro #DeveloperTools #AICoding #AgenticAI 🐈 This is a clear and conspicuous disclaimer that I am an AWS employee and all opinions are my own 🐈
-
-
KiroCrew 0.4.0 is out. The dashboard does more than watch. Real diffs, inline editing, and a file tree turn it into a full editor. → Windows and Linux now have signed installers, desktop automation on Windows, matching resource ceilings on both → More messaging channels catches up to Slack, plus WhatsApp, iMessage, and Feishu Secrets also move into an encrypted vault instead of sitting in your config file. CLI startup is 2x faster, and boot uses ~875MB less memory. What would make you actually work inside the dashboard instead of just checking on your agents? Full breakdown 👉 http://spr.ly/6040B1fW2A
-
Are AI coding agents actually getting better? Six months of diagnostics data say yes, with caveats. Kiro introduced Language Server Protocol (LSP)-based diagnostics directly into the coding agent from the very beginning of Kiro's agent architecture. The latest analysis of Kiro diagnostics data shows how models use LSPs and static analysis as a feedback loop to catch type errors, unresolved imports, undefined symbols, hallucinated properties, and more. The data suggests that newer models are not just making fewer errors; they are also checking broader sets of files and evolving in how they self-correct. Full detailed analysis here 👉 http://spr.ly/6042B1x15G
-
-
The Kiro IDE keeps evolving, and who better to walk us through the latest than one of the engineers building it? We're interviewing Al Harris, Principal Engineer on the Kiro team, as he shares what's new, what's improved, and how Agent Focus Mode reimagines the IDE experience with conversation at the center of your workflow, so you can focus on directing the agent instead of editing code line by line. Bring your questions live.
What's new in the Kiro IDE with Principal Engineer Al Harris
www.linkedin.com
-
System prompt behavior is hard to predict. An instruction that looks fine in isolation can break in cases nobody anticipated. So we built an LLM-judge pipeline that scores real Kiro conversations on a repeating loop: • Diagnose complaint patterns • Design targeted fixes • Test in cohorts • Evaluate with the same rubric every time 27 candidates screened across safety, verification, tone, and behavior. On Kiro CLI, behavioral quality issues dropped 32%. Full breakdown → http://spr.ly/6043B1QlG3
-
-
GPT‑5.6 🤝 Kiro GPT‑5.6, now available in Kiro, advances price-performance for developers. On Terminal-Bench 2.1, it completed successful tasks in Kiro at an 82% cost reduction, in under 4 minutes, with a 74% pass rate. Engineering rigor, quality, and intelligence. That combination is what efficiently turns AI coding sessions into finished work faster instead of requiring another round of prompting. Full breakdown → http://spr.ly/6041B1swhh