Plutonic Rainbows

Robert Aiki Aubrey Lowe’s album Power unfolds as a sculptural sound environment where voice, modular synth, and ambient recordings merge into a single, living organism. Lowe’s voice slides effortlessly between breath, chant, overtone, and electronic shimmer, interweaving with analog drones to build a ritual-like ambience. At its core is the space 'in between the notes' — a deliberate embrace of silence and nuance that allows the listener to become part of the unfolding sonic ritual .

This is not music to simply listen to, but music to inhabit. Layers of modular tones form an architectural soundscape, Lowe’s vocal harmonics act as both anchor and wandering specter, and field recordings introduce the presence of a broader world in motion. The album recalls the patiently nuanced, melancholic logic of his earlier work (like Kulthan), yet Power deepens that legacy into an immersive act of transformation. It’s an atmosphere you enter lightly but leave fundamentally changed — an experience of sound as living energy, at once intimate, uncanny, and profoundly ritualistic.

Available on Boomkat.

The new Supabase integration transforms prompt management from a manual, error-prone process into an automated, centralized system that ensures consistency across all AI interactions. By storing the master prompt_rules.xml file in a Supabase storage bucket, I can update the AI behavioral rules once through the web dashboard and have those changes automatically propagate to all local prompt templates through an intelligent synchronization system.

This eliminates the need to manually edit multiple prompt files, reduces the risk of inconsistent AI behavior, and provides a single source of truth for all AI interaction guidelines. The system leverages Supabase's robust storage infrastructure with proper authentication and access controls, while the smart sync mechanism only downloads and applies changes when rules have actually been modified, optimizing both performance and API usage.

I finally added this to my Strangelove collection.

There is a moment, just before the sun slips beneath the Pacific horizon, when the ocean reveals its truest nature — not just in color, but in feeling. Silence the Sea captures that rare synesthetic space where light, scent, and sound collapse into a single impression: vast, resonant, and eternal.

Opening with a breath of salted skin and sun-warmed minerals, this fragrance immediately conjures the chill of deep water and the glint of sunlight caught in surf. You smell the ocean not as a postcard cliché, but as a living, moving force — dark blue and fathoms deep. There’s a physicality to it, like the pressure you feel below the surface, or the echo of whales sounding in distant canyons. It’s not loud, but it is profoundly present.

Ambergris — the soul of this perfume — lends both texture and myth. It’s the scent of ancient marine stories, borne on tides and lifted by the wind. Here, it doesn’t dominate but rather shimmers quietly beneath crystal-clear waves, like something glimpsed just out of reach. Each breath seems to pull you further from shore, toward the open, unknown deep.

Yet, for all its abyssal resonance, Silence the Sea is illuminated by a kind of golden-hour light — that Pacific glow when the surf catches fire and the salt hangs softly in the air. There’s a skin-like warmth that rises over time, as if the sea has kissed you and then dried in the sun. It’s intimate, but never cloying; haunting, but never heavy.

This is not a perfume for those who like easy resolutions. It is a reverie of oceans — clean, deep, and unspeakably vast. Wearing it feels like stepping out of time, standing still in the surf, while the whales sing below and the horizon blazes above.

With Claude Code’s assistance, I built a React-based app featuring a simple countdown. Claude handled the entire process — developing the app, installing all necessary packages and dependencies, pushing it to GitHub, and connecting it to Vercel.

I also added a system, light and dark mode to the application. Again, Claude Code essentially carried out all the work.

Finally, I had Claude Code configure Sentry to monitor the blog in real time, with the added ability to query Sentry through their MCP server.

Yesterday, I spent some time revisiting Roja Parfum’s Lost In Paris and Dior’s Bois Talisman. Between the two, Bois Talisman left the weakest impression — the mid and dry-down stages have a peculiar plastic-like quality. Lost In Paris performed slightly better, though it felt somewhat linear, and I couldn’t detect the butter accord. For a fragrance in their prestige collection, it is something of a disappointment. Some are saying this is the best gourmand ever. I cannot say I agree.

In 2019, Cosey Fanni Tutti reignited her solo career with her first album TUTTI in almost four decades, and she has kept the flame diligently burning ever since. Returning with 2t2, she charges all sounds at her disposal into an arsenal of constantly morphing musical landscapes. In the album’s first half, all elements fluctuate around a driving, rhythmic core, with thrumming, pulsating synth lines slowly electrifying as vocals become abstract digital chants, while the second half is dressed in more obsidian tones, drifting from the electropop structures into something more formless and immersive.

Dior released this fragrance at the end of January. I’m still debating whether to add it to my collection — would it be redundant? I’m not sure. I’ll have to revisit my sample and see.

OpenAI drop the price of o3 by 80%. That's good news for people using the API. At this point, it really is very competitive with everyone trying to beat everyone else.

OpenAI o3 is the company's most powerful reasoning model that pushes the frontier across coding, math, science, visual perception, and more. It sets a new SOTA on benchmarks including Codeforces, SWE-bench (without building a custom model-specific scaffold), and MMMU. It’s ideal for complex queries requiring multi-faceted analysis and whose answers may not be immediately obvious. It performs especially strongly at visual tasks like analyzing images, charts, and graphics. In evaluations by external experts, o3 makes 20 percent fewer major errors than OpenAI o1 on difficult, real-world tasks — especially excelling in areas like programming, business/consulting, and creative ideation. Early testers highlighted its analytical rigor as a thought partner and emphasized its ability to generate and critically evaluate novel hypotheses — particularly within biology, math, and engineering contexts.

Updated: OpenAI have now added o3 Pro for users that are willing to pay $200 a month.

Reading a variety of books today.