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
7d agoGuides
AI Guides for Builders
How-to content for builders, indie hackers, and AI engineers. Less theory, more shipped code.
Machine Learning 
Tutorials Intermediate
Context Compaction: Run Claude Agents Past 200K Tokens
Use Anthropic's compact-2026-01-12 beta so long agentic loops survive past the 200K context window.
11 min read·Kodetra Technologies
8d ago
Tutorials Intermediate
Claude Task Budgets: Cap Agent Token Spend in Python
Give Claude agents a token countdown so long agentic loops finish gracefully instead of burning your bill.
9 min read·Kodetra Technologies
11d ago
Tutorials Intermediate
Claude Web Search Filtering: Cheaper Cited Agents in Python
Build cheap, cited research agents on Claude Sonnet 5 using the web search tool and dynamic filtering.
9 min read·Kodetra Technologies
11d ago
Tutorials Intermediate
Claude Managed Agents: Build a Long-Running Agent in Python
Stand up an autonomous agent with sessions, streaming events, and webhooks - no agent loop required.
9 min read·Kodetra Technologies
12d ago
Tutorials Intermediate
Claude Sonnet 5: Adaptive Thinking and Effort in Python
Master Sonnet 5's on-by-default thinking and the effort knob to cut cost and latency.
9 min read·Kodetra Technologies
13d ago
Tutorials Intermediate
Migrate to Claude Sonnet 5: Effort and the Tokenizer Trap
Hands-on Python guide to Sonnet 5's adaptive thinking, effort levels, and the 30% tokenizer trap.
10 min read·Kodetra Technologies
13d ago
Tutorials Intermediate
Claude Advisor Tool: Pair Haiku With Opus in Python
Let a cheap executor model consult a stronger advisor mid-task in one Messages API call.
11 min read·Kodetra Technologies
22d ago
Tutorials Intermediate
LLM Failover in Python: Survive a Model Going Dark
Build a provider-agnostic LLM failover client in Python that survives outages and model removals.
9 min read·Kodetra Technologies
22d ago
Tutorials Intermediate
Loop Engineering: From Prompts to Verified Agent Loops
Build a plan-act-verify agent loop with an external check, retry budget, and clear stop rules.
9 min read·Kodetra Technologies
27d ago
Tutorials Intermediate
Fable 5 Prompt Caching: Slash 1M-Token Codebase Costs
Reuse a huge codebase prefix across every Fable 5 call and pay ~90% less.
8 min read·Kodetra Technologies
28d ago
Tutorials Intermediate
Fable 5 Effort: Cut Thinking Token Costs in Python
Claude Fable 5 always thinks. Use effort, display and max_tokens to control reasoning cost.
9 min read·Kodetra Technologies
29d agoPage 1 of 2