I examined Claude Code, Codex, and Antigravity on an actual electronics mission — just one really completed

AI coding brokers are a dime a dozen in 2026, however not all of them are value your time. Totally different fashions reply in another way to issues, and a mannequin that may create a flawless web site in minutes may not be pretty much as good at writing bash scripts. Now there are particular tasks the place I will not vibe code, however engaged on smaller electronics tasks is the right place for these fashions to shine.

So I put Claude Code, Codex, and Antigravity to the take a look at in an actual electronics mission — a easy dashboard with an ESP32, an OLED show, and a temperature sensor that may do a number of issues directly. And whereas all three fashions can get you a working prototype, you’d not need to blindly choose your coding assistant.

A easy dashboard was a severe take a look at

Constructing an ESP32 gadget that combined APIs, sensors, shows, and Residence Assistant

Should you’re working with bigger codebases or monorepos, Claude Code’s reasonably restricted context window and utilization limits generally is a drawback. I solved Claude Code’s reminiscence drawback with a Postgres database, however always clicking a tiny icon to see how a lot of my utilization quota is left was irritating.

I made a decision to repair this by making a small devoted dashboard that confirmed my utilization limits, so as an alternative of clicking round or typing in instructions, all it takes is a look to see the place I am at. I additionally determined so as to add a temperature sensor so the dashboard may also present my workplace’s ambient temperature and humidity, which I can pipe into my Residence Assistant dashboard to arrange automations for my air conditioner.

It is a comparatively easy construct. All you want is an off-the-shelf ESP32 — no threat of selecting the unsuitable ESP32 board right here — a DHT22 temperature sensor, a generic monocolor OLED display (I am utilizing an SH1106 show, however some other show works superb), and a easy push button to cycle by way of the knowledge on the show. That is it.

I spent a while in Blender to give you a fast enclosure for the mission, and a fast 3D print later, I had every thing I wanted. Now it was time for the coding brokers to get to work. All three brokers have been supplied the identical immediate as follows:

I need to make an ESP32-based dashboard that does three issues:

1. It ought to present my Claude plan restrict utilization
2. It ought to present the temperature and humidity studying from an connected DHT22 sensor and in addition report it to my Residence Assistant occasion.
3. It ought to present me the climate and temperature exterior in New Delhi. I’ve an ESP32 Devkit V1 board, a DHT22 sensor, and an SH1106 128×64 monochrome OLED.

There’s additionally a 4-pin push button that can be utilized to cycle modes between the three functionalities and replace the show accordingly.

The aim was to achieve a working prototype assembly all three targets in as few and as easy prompts as attainable. All three brokers would take a couple of immediate to get the mission proper, however they took considerably totally different approaches to get the job completed.

Model

Espressif

Connectivity Options

Wi-Fi, Bluetooth

ESP32 is a low-cost microcontroller with built-in Wi-Fi and Bluetooth, broadly used for IoT tasks and DIY electronics.

Antigravity moved quick

Spectacular pace, however extra planning than doing

First up on the listing was Google’s Antigravity operating Gemini 3.1 Professional on Excessive effort, and it’s quick, simply not in a method you’d anticipate. As quickly because the mannequin acquired its immediate, it searched the web for test Anthropic API plan restrict utilization and offered an implementation plan with some follow-up questions.

I began all three brokers on the similar time, and whereas Codex and Claude Code each got here up with the primary model of a program able to be flashed to the ESP32, Antigravity insisted on my approving its implementation plan earlier than it wrote a single line of code. After I cleared up its doubts, it got here up with yet one more implementation plan and clearly acknowledged that it will not be capable to fetch my Claude Professional subscription’s utilization limits, as there is no such thing as a official API from Anthropic that allows this performance. The temperature information from the sensor could be despatched over to Residence Assistant through its built-in REST API, which is the better approach to do it until you are utilizing one thing like ESPHome.

I then piped in Hermann Björgvin’s Clawdmeter GitHub repo, which is an identical mission primarily based on the Waveshare ESP32-S3-Touch-AMOLED-2.16 board, to push it alongside. It was then that the mannequin was ready to determine that it may use the OAuth entry token generated by the Claude CLI and ship it to a particular beta API endpoint to fetch the bounds. Thoughts you, Antigravity nonetheless hadn’t produced a single line of code thus far. Should you’re of the opinion that Antigravity is not able to tackle Claude Code and Codex, this does nothing however reinforce it.

As soon as I accepted the third and last motion plan, it created a PlatformIO mission that labored for roughly two minutes to supply a collection of recordsdata for me to sort in my Wi-Fi credentials and OAuth token individually from the principle supply code, all able to be flashed on the ESP32. That is the fitting strategy when it comes to safety, however for a tool that’ll by no means go away your native community, it is unnecessarily complicated. To not point out that should you’re a newbie to growth boards just like the ESP32 and the Arduino IDE, seeing a bunch of recordsdata if you’re solely anticipating one could be overwhelming.

OS

Home windows, macOS, Linux

Developer(s)

Google

Engine

Gemini 3.1 Professional / Gemini 3 Flash

Google Antigravity is an agent-first built-in growth atmosphere (IDE) designed for autonomous software program growth. Constructed on a modified Visible Studio Code basis, it permits a number of AI brokers to independently plan, code, and take a look at purposes.

Codex by no means actually understood the project

The errors that stored it from producing a usable end result

OpenAI’s Codex, utilizing GPT 5.5 on Excessive effort, wasted no time and produced two YAML recordsdata in about three minutes from receiving the immediate, whereas stepping into a very surprising path. It used ESPHome to configure the ESP32 as a sensor that may natively present up in my Residence Assistant occasion. This may work finest if I have been working with simply the DHT22 sensor to create a devoted temperature sensor for my workplace, however for a board dealing with two different duties, it wasn’t fairly the fitting strategy.

Codex additionally identified that Anthropic had no official API for fetching Claude Professional utilization limits. Apparently, it did not let that hurdle cease it. Codex created a Residence Assistant input_number referred to as input_number.claude_plan_usage_percent to basically manually set a quantity I replace myself (or through an automation) to point utilization. Offering the choice to create an actual API-backed HA sensor if I had entry to Anthropic’s Console group or admin key. I do not.

In my follow-up, I supplied the GitHub hyperlink to the identical Clawdmeter mission I gave to Antigravity. It understood the strategy, however as an alternative of ditching ESPHome, it doubled down on the identical path of utilizing manually set numbers as earlier than and prompted that the following step could be including a Clawdmeter-style bridge that writes actual Claude Code utilization into the helper lessons it created beforehand. It did, nonetheless, spit out a reasonably clear wiring diagram for connecting all of the parts.

Now this strategy can work, however it requires loads of workarounds and suboptimal practices that add complexity for no good motive. It appears as if Codex determined to concentrate on the temperature sensor a part of the instruction (doubtless as a result of it’s higher documented) and bent every thing else to work in that ecosystem as an alternative of adjusting the construction to raised go well with the duty at hand.

It will take fairly a little bit of forwards and backwards to get the prototype working as anticipated. Actually, it will’ve been sooner for me to put in writing the code myself as an alternative of losing time down the ESPHome rabbit gap that Codex was so persistent about.

OS

Home windows, macOS, Linux

Developer

OpenAI

Value mannequin

Free, paid plans obtainable

Codex is OpenAI’s software program engineering agent that may learn, write, debug, and modify codebases utilizing pure language directions to automate growth duties.

Claude Code took its time, however it acquired the duty proper

Slower iterations, higher reasoning, and code that truly match the duty

Claude Code utilizing Sonnet 4.6 on Excessive effort was the slowest of the three brokers proper out of the gate. Primarily as a result of I used to be always exhibiting a mannequin overloaded warning — indicating excessive server load. This can be a server-side subject that you would be able to’t do a lot about, and it is the primary time I am seeing this within the a number of months I have been utilizing Claude Code, so I made a decision to let it slide.

After some pondering, Claude produced a single INO file (utilized by the Arduino IDE) that did all three duties proper from the primary model. Not like Antigravity’s refusal or Codex’s delusion, Claude Code solved the utilization restrict fetching drawback by sending POST /v1/messages with max_tokens:1 to Haiku, returning the headers that present requests and tokens remaining within the present window. It was extraordinarily clear about what the outcomes would and wouldn’t present, and my month-to-month token consumption towards my Claude Professional subscription wasn’t included.

Identical as earlier than, I threw in a hyperlink to the Clawdmeter mission to assist the agent alongside, and it instantly tailored. Claude Code up to date the Arduino sketch with the brand new headers, spit out a preview of what could be proven, and a single INO file able to be flashed to the ESP32, together with directions on the place to seek out the Claude OAuth token.

Just like Antigravity, Claude Code used Residence Assistant’s built-in REST API to ship over temperature information from the sensor, however crucially, it additionally identified that the mixing will not persist by way of Residence Assistant server restarts until the board posts information once more earlier than the restart is full. It advised utilizing MQTT for a cleaner integration, which might require some further setup on the Residence Assistant entrance.

Developer

Anthropic PBC

Value mannequin

Free, subscription obtainable

Claude Code is a complicated synthetic intelligence coding agent developed by Anthropic. Constructed on Constitutional AI ideas, it excels at complicated reasoning, subtle writing, and professional-grade coding help.

Just one agent crossed the end line

Which mannequin finally delivered a working dashboard and why

For my goal, Claude Code was the mannequin that produced the very best outcomes. It gave me a clear wiring diagram and a single INO file to work with, which lowered mission complexity. On a tool that is going to sit down on my native community, there is not any motive so as to add further recordsdata for OAuth tokens or Wi-Fi credentials.

Moreover, since I take advantage of Claude extra, it already had my board and sensor stock saved in reminiscence, it knew the native IP deal with my HA occasion lives on, and it wired all of that instantly into the code, giving me fewer issues to edit earlier than flashing the ESP32. You would possibly deal with Clade Code as an agent, however it may be a military with the fitting context. Should you’ve been utilizing Codex or Antigravity for some time, the identical would possibly apply to you.

Antigravity was a detailed second. It tried to do every thing, and it tried to do it proper, however added pointless complexity, and the code needed to be manually patched and debugged earlier than it will work. Antigravity additionally did not present a wiring diagram, which means I needed to look by way of the code to determine wire up my {hardware}.

Final however not least, Codex made the fitting name with ESPHome, however solely within the context of the dashboard primarily being a temperature-sensing station. Regardless of offering steerage on fetch utilization limits, it continued down the ESPHome path, including pointless complexity and forcing me to arrange the entire thing in YAML when it will’ve been a lot simpler to flash the board through Arduino for this sort of activity. You’ll be able to ask the agent to divert it is strategy and create an Arduino sketch that works equally to the Antigravity and Claude Code strategy, however that is one other back-and-forth spent troubleshooting and fixing mission structure.

When pace issues, when accuracy issues, and the place every coding agent matches

As I discussed earlier than, all three approaches would work, however some would require extra tinkering than others. One software is not universally higher than the others; it is that embedded firmware exposes the hole between instruments that generate code and instruments that truly perceive the issue. Relying on which software you most frequently use and what strategy you need to take, your outcomes will fluctuate.

Antigravity is quick and spectacular with the fitting context. Codex falls again to raised documented strategies, even when they’re unsuitable for the duty at hand. However on a mission the place totally different subsystems must cooperate on constrained {hardware}, Claude Code’s slower, extra deliberate, and general easier strategy produced the one end result I may really flash to the board and begin utilizing immediately.