WeAreDevelopers Logo WeAreDevelopers
Sign in
For You LIVE
World Congress 2026 Europe • Jul 10, 2026 • Session details

Completing the Feedback Loop

Nimrod Kor

LLMs can't grade their own work. Move beyond basic prompt responses by building multi-layered agent workflows. Learn how external grounding drives true autonomous task completion.

Pause
Mute Enter Fullscreen
#1 about 2 min

Understanding the necessity of AI agent feedback loops

An overview of why agents often fall short and how providing appropriate context guarantees success.

#2 about 2 min

Expanding AI agents for code review and security scanning

Specialized software agents automatically manage code reviews, scan for security vulnerabilities, and handle repository merges.

#3 about 1 min

Essential components of the reason and act agent loop

Core agent architectures depend on language models paired with custom tools and the standard reasoning loop.

#4 about 1 min

Defining tools and response formats in application code

Implementing code frameworks requires explicitly defining tools, system prompts, and structured response classes for reliable outputs.

#5 about 2 min

Designing effective custom tools with descriptions and string outputs

Detailed phrasing within a tool's description directs models appropriately while formatting complex results as strings prevents constraint errors.

#6 about 3 min

Structuring thinking and verdicts in code merging agents

Ordering logical operations carefully ensures automated merge systems consider all evidence before rendering final decisions.

#7 about 2 min

Differentiating bounded agent loops from continuous agentic loops

While standard loops process bounded single tasks, an agentic loop continuously monitors environments to trigger new automated responses.

#8 about 2 min

Breaking large requirements into manageable orchestrator tasks

Separating massive engineering requests into product requirement documents empowers automated orchestrators to securely manage isolated execution threads.

#9 about 2 min

Implementing automated spec reviews across complex application environments

Integrating browser endpoints and external workflow trackers introduces broader context for evaluating sprawling new feature requests.

#10 about 2 min

Overcoming spec fatigue and hallucinated requirements in single agents

Dumping massive specification lists directly into isolated models causes missed constraints and hallucinated logic blocks.

#11 about 3 min

Scaling validation with map reduce workflow architectures

Distributing requirement verifications into parallel checks prevents duplicate evaluations while dramatically accelerating completion times.

#12 about 1 min

Optimizing efficiency by matching specific models to workflow steps

Sourcing heavy models for context extraction while matching lightweight models to isolated validations minimizes costly token usage.

#13 about 2 min

Validating code outputs against actual user interface reflections

Correlating logical structural modifications directly with integration tools reveals the disconnects within positive self-evaluations made by language models.

#14 about 2 min

Strategizing comprehensive output validation and tool optimization

True feature validation demands custom task dissections paired with concrete external test components to guarantee proper code integrations.

#15 about 2 min

Closing deployment feedback gaps with reality based behavioral checks

Exposing code bases to authentic behavior queries protects deployment confidence against severely flawed model self-grading parameters.

Matching moments

1:56 min

Introduction to evaluating AI code review agents

Sofia Rest Sofia Rest · World Congress 2026 North America

2:06 min

Rethinking team structures around AI agent capabilities

Mike Mike · World Congress 2025

1:19 min

Turning repetitive developer tasks into automated AI agent skills

Markus Eisele Markus Eisele · World Congress 2026 Europe

2:09 min

Creating reliable feedback loops and automated testing for AI

Clemens Wasner Clemens Wasner +4 · World Congress 2026 Europe

2:40 min

Reviewing live performance of self-correcting AI engineering agents

Ingo Eichhorst Ingo Eichhorst · World Congress 2026 Europe

10:17 min

Discussion on AI hallucinations and practical developer workflows

Akmal Chaudhri Akmal Chaudhri · LIVE

More from World Congress 2026 Europe

Building the Nervous System of AI - Michael Kagan (NVIDIA) thumbnail
29:17
3 months ago

Building the Nervous System of AI - Michael Kagan (NVIDIA)

Michael Kagan, Sead Ahmetović

Five Things in Tech that Matter Now - WeAreDevelopers World Congress 2026 Closing Keynote thumbnail
19:41
3 months ago

Five Things in Tech that Matter Now - WeAreDevelopers World Congress 2026 Closing Keynote

Christian Heilmann

The Sound of Your Secrets: Teaching Your Model to Spy, So You Can Learn to Defend thumbnail
32:06
3 months ago

The Sound of Your Secrets: Teaching Your Model to Spy, So You Can Learn to Defend

David vonThenen

Super scaling for the Super Bowl: How to survive 30 million users hitting your backend in 30 minutes thumbnail
35:13
3 months ago

Super scaling for the Super Bowl: How to survive 30 million users hitting your backend in 30 minutes

Irina Branovic

Running Secure Life Science Research at Scale using Hybrid GPU HPC and Kubernetes 🧬 thumbnail
24:23
3 months ago

Running Secure Life Science Research at Scale using Hybrid GPU HPC and Kubernetes 🧬

Jeremy Murray

Angstfreude - AI and Corporate Culture - The Thrill and the Threat thumbnail
30:26
3 months ago

Angstfreude - AI and Corporate Culture - The Thrill and the Threat

Axel Ebert, Karin Krobath

Related videos

You don't need to write the code. You need to become a verification architect and prove it's correct thumbnail
27:43
World Congress 2026 Europe • 3 months ago

You don't need to write the code. You need to become a verification architect and prove it's correct

Guillaume Moigneu

Teaching an LLM to review code … like a Senior Engineer! thumbnail
32:26
World Congress 2026 Europe • 3 months ago

Teaching an LLM to review code … like a Senior Engineer!

Kesha Mykhailov

What Production Knows: Closing the Loop Between AI Agents and the Systems They Build thumbnail
26:33
World Congress 2026 Europe • 3 months ago

What Production Knows: Closing the Loop Between AI Agents and the Systems They Build

May Walter

5 things I wish I hadn’t done building my AI agent thumbnail
20:16
World Congress 2026 Europe • 3 months ago

5 things I wish I hadn’t done building my AI agent

Shachar Azriel

Agents Can't Iterate Against Tests That Lie thumbnail
29:46
World Congress 2026 North America • 11 days ago

Agents Can't Iterate Against Tests That Lie

Rocky Warren

Beyond Vibe Coding: Using CLI Tools as Your AI Counterpart in Pair Programming thumbnail
30:03
World Congress 2026 North America • 12 days ago

Beyond Vibe Coding: Using CLI Tools as Your AI Counterpart in Pair Programming

Ron Veen

Related articles

From learning to earning

Jobs that call for the skills explored in this talk.