All Articles
Fixing IME Input Issues in Antigravity: Dropped Composition, Premature Commits, and Keybinding Clashes
If you type Japanese, Chinese, or Korean inside Antigravity, AI inline completion and keybindings sometimes eat your IME composition before you can commit it. Here is a pattern-by-pattern fix covering macOS, Windows, and Linux.
Antigravity × Local LLM (Ollama / LM Studio / LM Link): A Production Connection Playbook
Getting Antigravity to connect to Ollama, LM Studio, or LM Link is the easy part. Running on that setup for eight hours a day, week after week, surfaces disconnects, model-swap hangs, stale sessions after lunch, and VRAM pressure from other processes. This playbook covers hardening the connection layer, picking the right backend for the task, and designing the fallback path for when local goes silent.
Antigravity Retry Stuck in a Loop? A Triage Guide That Actually Breaks It
Pressing retry in Antigravity feels like it should eventually work, but sometimes the same failure keeps coming back with only tiny variations. This guide names the three modes the retry loop falls into, walks through a triage flow, and gives you a rule of thumb for when to stop retrying and start intervening.
Production Multi-Agent Systems with Antigravity AgentKit 2.0: Patterns, Failure Modes, and What the Demos Don't Show
AgentKit 2.0 makes multi-agent systems look effortless in demos, but running them in production is a different problem. This guide covers Planning vs. Fast mode, three real orchestration patterns, and the failure modes — infinite loops, cost blowouts, prompt injection — that bite on day one.
Semantic Caching for LLM Responses in Antigravity — Estimating Your Own Savings Before You Build It
Building a semantic LLM response cache with Antigravity, pgvector, and Gemini. Covers migrating off the retired text-embedding-004, normalizing truncated vectors, and a formula that turns hit rate and unit-cost ratio into an actual savings estimate.
Using Antigravity's Retry Feature Wisely — A Smarter Way to Resume Failed Agent Runs
Antigravity's Retry button is not a reroll. This guide explains when retry actually helps, how to prepare context before retrying, and when you should stop retrying and start a fresh session.
Designing Antigravity Browser Agents That Don't Go Flaky — Taming DOM Drift and Timing
A browser-agent task that passed yesterday trips today, then succeeds on retry, then fails again. Here's how I've been breaking down the causes of flaky Antigravity browser agents in production and addressing them at the design level.
Observing Antigravity AI Agents with Langfuse — A Practical Setup Before You Ship
Before you push an Antigravity AI agent to production, wire up Langfuse so you can actually see traces, token spend, and cost. A hands-on guide with real Python code and lessons from the field.
Keeping the Antigravity Python API Stable in Production — Retries, Timeouts, and Circuit Breakers That Actually Work
A deeply practical guide to keeping Python services built on the Google Gen AI SDK alive under real traffic. We cover retry, timeout, circuit breaker, rate limit, and cost budgeting patterns with runnable code from an Antigravity workflow.
Build a CSV Analysis Agent in Antigravity with pandas and Matplotlib
Walk through building an Antigravity agent that analyzes CSV files from natural-language instructions. Covers the pandas tool design, common pitfalls, and safety practices for production use.
Prompt Injection Defense in Antigravity — A Production Security Playbook for LLM Apps
Build a four-layer prompt injection defense inside Antigravity with working Python. Includes measured heuristic recall across eight obfuscations and the normalization bug that breaks detection for non-Latin scripts.
Memory Budget Design for Embedding Gemma 4 in Mobile Apps
When embedding Gemma 4 into a mobile app, peak memory during inference — not the model file size — becomes the real bottleneck. A memory budget design based on measured values, plus a device-tier switching strategy.