Audio Serial Storage
A tape-style data encoder/decoder using FSK audio, compression, Reed–Solomon error correction and CRC integrity checks. It is inspired by 8-bit cassette storage and low-bandwidth amateur-radio data links.
I am a Senior Technical Account Manager at Red Hat in Finland, helping organisations make practical use of OpenShift, Kubernetes, OpenShift AI, RHEL AI and large language models.
LinkedIn · GitHub · Linux/OpenShift blog · Amateur radio · Music · Photography
Loading amusing thought...
My current work centres on enterprise platforms and the transition from experimentation to useful, supportable AI services. I work with customers and technical teams on OpenShift and Kubernetes architecture, application platforms, automation, operational readiness, AI/ML workloads, model serving and the practical adoption of generative AI.
Earlier work gives me a broad systems foundation across UNIX/Linux, enterprise infrastructure, data migration, virtualisation, automation, monitoring, security analysis, databases and HPC. The older technologies still inform how I approach reliable modern platforms, but my present focus is cloud-native and AI-enabled systems.
Alongside my professional work, I build small systems and proof-of-concepts that explore unusual interfaces, local AI, knowledge bases, language learning and specialist communities.
A tape-style data encoder/decoder using FSK audio, compression, Reed–Solomon error correction and CRC integrity checks. It is inspired by 8-bit cassette storage and low-bandwidth amateur-radio data links.
A Python acoustic timegrapher for mechanical watches and pendulum clocks. It uses microphone or piezo input, signal processing, beat-error measurement and trend graphs as a software alternative to dedicated timing equipment.
A WordPress plugin that connects a site to an OpenWebUI backend, bringing an LLM-powered chat experience into a web application.
A command-line experiment for executing shell commands with the assistance of a large language model.
An independent amateur-radio dashboard for FreeDV activity, with real-time visualisation and local LLM-powered summaries of traffic, band openings and propagation. I describe the project and its purpose in the M0SPN/OH3SPN technical blog.
A broader platform for specialist markets, focused search and LLM-assisted information discovery. Current projects include Ribbon Radar, an EGL/Lolita fashion sales aggregator incorporating LLM experiments, and Kieli Kone, a language-learning tool incorporating the Finnish YKI requirements.
Music and old-school computing remain important parts of what I do. My Amiga connection goes back to the demoscene: I was involved with Slipstream, an Amiga/PC/CPC/C64 demogroup, and continue to explore Amiga software through AmiTCP_NG. I also maintain an interest in Amiga.org, AmigaOS and the wider retro-computing community, including the Swindon Museum of Computing.
Current musical projects include Coherer, Cybernet 1202, Town and County and Pennycross Coven.
Bitstream Podcast showcases retro gaming and demoscene audio, including music from classic computers, games and demoparties.
Earlier projects include Rammlied, The Beauty of Grey and Caffeine Kill.
Photography is another long-running interest, spanning both film and digital work. I share selected images through my Flickr stream, Pixelfed stream and Mastodon photography profile.
My Only Linux Makes It Possible blog contains technical notes and articles about Linux, OpenShift, AmigaOS, local LLMs and related experiments.
Other enduring interests include Amiga.org, the Swindon Museum of Computing, Akaa Hacklab and Tampere Hacklab.
I operate as M0SPN/OH3SPN. My M0SPN/OH3SPN technical blog covers digital voice, FreeDV, antennas, propagation, operating and radio experiments; recent work includes FreeDV Reporter+.
Other relevant links include the Radio Society of Great Britain and Suomen Radioamatööriliitto.
Radio, electronics and low-bandwidth digital modes continue to influence my software experiments.
Email: steve at netting dot org dot uk · LinkedIn · GitHub
Amateur radio: M0SPN / OH3SPN. HF/VHF/UHF voice, CW and PSK by arrangement.