Tutorials Advanced
Steer GPT-6 Astra Mid-Turn Without Losing Its Work
Redirect a running Astra agent over WebSocket with response.steer, keeping finished work.
13 min read·Kodetra Technologies
TodayGuides
AI Guides for Builders
How-to content for builders, indie hackers, and AI engineers. Less theory, more shipped code.
Tutorials 
Tutorials Intermediate
Qwen3.8-Flash-Next: Preserved Thinking, Cheaper Agents
Qwen keeps reasoning across every turn by default. Exploit it, or pay for it.
11 min read·Kodetra Technologies
9d ago
Tutorials Advanced
Claude Browser Toolset: Ref Targeting Over Pixels
Wire browser_toolset_20260801 to Playwright: ref targeting, batched turns, halt-on-failure.
15 min read·Kodetra Technologies
15d ago
Tutorials Intermediate
MRTR: Human-in-the-Loop MCP Tools Over Stateless HTTP
Pause an MCP tool mid-call to confirm with the user, on the new stateless 2026-07-28 protocol.
10 min read·Kodetra Technologies
26d ago
Tutorials Advanced
Swap Tools Mid-Chat on Claude Opus 5, Keep the Cache
Add or drop Claude Opus 5 tools between turns without invalidating your prompt cache.
12 min read·Kodetra Technologies
Jul 28
Tutorials Intermediate
Kimi K3 Partial Mode: Prefill Replies to Force Output
Prefill Kimi K3's assistant turn to lock output shape, force clean JSON, and steer tone.
9 min read·Kodetra Technologies
Jul 21
Tutorials Intermediate
Kimi K3 Vision: Extract Structured Data From PDFs in Python
Use Kimi K3's native vision and JSON schema output to turn messy PDF pages into clean, typed data.
9 min read·Kodetra Technologies
Jul 18
Tutorials Intermediate
GPT-5.6 Persisted Reasoning: Reuse Thinking Across Turns
Use reasoning.context to reuse GPT-5.6's chain of thought across turns and cut redundant tokens.
9 min read·Kodetra Technologies
Jul 16
Machine Learning Advanced
Verifiers v1: Train Agents Past the Context Window
Ship a taskset, swap any harness, and turn compacted rollouts into real RL training samples.
13 min read·Kodetra Technologies
Jul 14
Machine Learning Intermediate
SWE-Together: Grade Coding Agents on Multi-Turn Sessions
Build a SWE-Together-style multi-turn coding-agent eval with an LLM user simulator in Python.
10 min read·Kodetra Technologies
Jul 8
Machine Learning Advanced
Read an LLM's Mind: Probe Hidden Thoughts Like J-Space
Train a linear probe on hidden activations and steer output, the method behind Anthropic's J-Lens.
9 min read·Kodetra Technologies
Jul 7
Tutorials Advanced
Harden Your Local MCP Server Against AutoJack RCE
AutoJack turned a local MCP server into RCE. Harden yours with origin checks, auth, and allowlists.
9 min read·Kodetra Technologies
Jun 24Page 1 of 2