majestic5762
Joined 4 years ago
Karma
85
Submissions
98
Member since
Jun 1
Recent submissions
Comment · 2 months ago
Comment · 3 months ago
This is my vision of the future.
To be transparent, I began by writing down my own thoughts, then used Sol to help rewrite them into a more publishable form. I spent considerably more time editing and refining the resulting draft.
I feel obliged to mention this because I can smell AI-generated crap from a mile away - and so can you. I have no interest in wasting anyone’s time with it. In this case, however, Sol did an excellent job of understanding my ideas and presenting them in a clear and digestible format.
Comment · 3 months ago
Just use GPT-5.6 Model Router from the Codex plugins
Comment · 2 years ago
They want to be the only ones spying
Comment · 2 years ago
Thoughts:
1) add ffmpeg wasm
2) use ffmpeg to detect scene changes
3) you can generate nice & short social media videos based on scene changes
Comment · 2 years ago
yeah, useless stuff portrayed as smart
Comment · 2 years ago
mykin.ai has the best memory feature i've tested so far. ChatGPT's memory feature feels like a joke compared to Kin's
Comment · 2 years ago
rewind.ai
Comment · 2 years ago
mykin.ai is building this with privacy in mind. Runs small models on-device, while large ones in confidential VMs in the cloud.
Comment · 2 years ago
I was thinking a couple of days ago about this concept of windowing for LLMs, but I lack the technical skills to implement it. Now Apple just published a paper on it. This is what I call synchronicity
Comment · 2 years ago
Yahoo Pipes was full of XSS vulns. I used to play with cookie grabbers
Remote: Yes
Willing to relocate: Yes
Résumé/CV: https://bit.ly/4wZMrTy
Email: [email protected]
Technologies: TypeScript, JavaScript, Node.js, Bun, React, Next.js, Vue.js, React Native, Expo, Fastify, NestJS, Hono, Swift, Kotlin, Go, Python, PostgreSQL, Prisma, Drizzle, Neon, Hasura, GraphQL, tRPC, WebSockets, RabbitMQ, Inngest, Mastra, LangChain.js, RAG, vector databases, embeddings, LLM agents, LLM evals, voice AI, LiveKit, ONNX, Transformers.js, speech-to-text, text-to-speech, on-device inference, CRDTs, distributed state, local-first architecture, offline-first applications, real-time synchronization, end-to-end encryption, SDK and CLI development, OAuth/OIDC, Better Auth, multi-tenant systems, Jest, OpenTelemetry, Grafana, Docker, Kubernetes, Nomad, CI/CD, GitHub Actions, Vercel, Fly.io, Cloudflare, Azure, AWS, GCP, Solidity, Pact smart contracts, zero-knowledge proofs, Konva.js, frontend performance, application security, and infrastructure security.
I've built AI products, developer platforms, mobile applications, distributed systems, and production infrastructure from early architecture through operation. Recent work includes an institutional financial-intelligence platform, LLM-powered ETL and evaluation pipelines, real-time voice agents, a local-first TypeScript framework, and a privacy-first personal AI product with on-device inference.
I can take ownership of a product or platform, from uncertain requirements to a reliable production system. I'm particularly useful for AI-native products, founding/lead engineering, developer tools and SDKs, local-first or real-time applications, mobile/web platforms, security-sensitive systems, and infrastructure that needs to scale. I can define the architecture, ship across the stack, establish evaluation and observability, improve engineering execution, and work directly with founders, clients, and product teams.