Luke Oliff · Blog
Writing.
Original posts, published first on this site. Imported posts live in the archive.
SearchAI (50)API Design (13)APIs (23)Career (6)Community (1)Developer Experience (52)Developer Relations (2)Engineering (124)Fun (1)Git (12)Industry (29)JavaScript (6)Latency (2)Models (2)News (17)Open Source (19)Opinion (35)Pricing (1)Security (10)Speech-to-Text (9)Streaming (4)Text-to-Speech (29)TIL (110)Voice Agents (7)Voice AI (74)
260 posts
2026
Three STT Strategies, One MarketOriginalBy June 2026 accuracy stopped deciding the STT market. Multilingual support, turn detection and platform integration became the real differentiators.Voice AI
Inside the Voice Agent Pipeline: STT, LLM, and Streaming TTSOriginalHow STT transcribes audio, an LLM generates responses, and streaming TTS speaks them back. A technical breakdown of the real-time pipeline behind voice agents inVoice AI
TIL: Flux natively knows when a caller finishes speakingOriginalDeepgram Flux turn detection replaces VAD and silence timeouts with model-native EndOfTurn events.Voice AI
Microsoft shipped 7 MAI models and AI hit $580BOriginalAI funding hit $581.7B in 2025. Microsoft shipped 7 MAI models. Apple chose privacy at WWDC. Deepgram kept shipping. Stories from a week that shifted AI.AI
Sunday roundup: afterthoughts on productivity and paceOriginalA short Sunday roundup on writing about ADHD and productivity, what Deepgram shipped this week, and what is coming next.Developer Experience
Six tools that power production voice agentsOriginalBuilding a production voice agent takes more than an API key. Here are six tools I reached for daily at Deepgram, from streaming STT to async Python.Developer Experience
Testing Voice AI Means Talking to Yourself in PublicOriginalBuilding voice AI means reading test sentences aloud in coffee shops, on trains, and in meetings you forgot to mute. It looks as ridiculous as it sounds.Voice AI
I'm not apologising for being productiveOriginalTwenty thousand more corrective messages by the age of ten is what an ADHD childhood costs. I am done apologising for being productive.Opinion
When SDKs Write Themselves: Voice API Code GenerationOriginalDeepgram swapped hand-rolled SDKs for spec-first generation with Fern, so five languages write themselves from one OpenAPI and AsyncAPI source.Developer Experience
TIL: Encoding Files as Base64 in the TerminalOriginalbase64 encodes any file into text-safe output. Embed small audio clips or images inline in JSON payloads without hosting them somewhere.TIL