AI agents, internal tools and automation for small B2B teams.

Three client systems described in architectural detail with the client’s name withheld, and two products of my own that you can open. No metrics appear here that I can’t source.

1. Support agent platformAn AI support agent across tickets, live chat and a first-party chat widget, with a learning loopNext.js 16 · AI SDK v6 · Upstash Redis + Vector · Claude, Gemini, OpenRouterAn AI support agent across tickets, live chat and a first-party chat widget, with a learning loopNext.js 16 · AI SDK v6 · Upstash Redis + Vector · Claude, Gemini, OpenRouterIn productionClient work
2. NEPSE CopilotA multi-account investing copilot for the Nepal Stock Exchange that grades its own callsNext.js 16 · Neon Postgres · AI SDK v7 · Gemini Live · TelegramA multi-account investing copilot for the Nepal Stock Exchange that grades its own callsNext.js 16 · Neon Postgres · AI SDK v7 · Gemini Live · TelegramLive · sign-in ↗Own product
3. Support IntelligenceA morning read of a support desk for people who don't read ticketsNext.js · Gemini 2.5 Flash · RedisA morning read of a support desk for people who don't read ticketsNext.js · Gemini 2.5 Flash · RedisIn productionClient work
4. offScriptA rehearsal partner for actors: two speech engines, a directing room, and a cue engine that knows the lineNext.js · Gemini TTS + ElevenLabs · Gemini Live · browser speech · IndexedDBA rehearsal partner for actors: two speech engines, a directing room, and a cue engine that knows the lineNext.js · Gemini TTS + ElevenLabs · Gemini Live · browser speech · IndexedDBLive ↗Own product
5. Community SignalTwo small internal tools that each replaced a weekly manual taskPython · SQLite · Claude / Next.js · monday.com GraphQL · GeminiTwo small internal tools that each replaced a weekly manual taskPython · SQLite · Claude / Next.js · monday.com GraphQL · GeminiInternalClient work

1. 01

   Scoping call

   A yes, a no, or a smaller first project.

2. 02

   Written spec and fixed price

   A spec you can hold me to, and a price that doesn't move.

3. 03

   Build, with weekly dry-run demos

   A working system you've watched rehearse.

4. 04

   Handover

   A repo, a runbook, and a system your team can run without me.

01

What does Mytrya do?

Mytrya builds AI agents, internal tools and data pipelines for small B2B software and service companies. It is a one-person practice run by Suraj Malla in Kathmandu. Every system described on this site was built and is run by him.

02

Which systems can you integrate with?

Anything with an API, a webhook, or a web page. Shipped work so far has connected Freshdesk, Freshchat, FastSpring, monday.com, Slack, Telegram, WordPress, Bettermode, MeroShare, three Nepali market-data sites, a broker's web terminal, Claude, Gemini and Gemini Live, OpenAI, OpenRouter, ElevenLabs, Neon Postgres, Upstash, Redis Cloud, Vercel Blob and Cloudflare. A system that isn't on that list is usually a day of adapter work, not a reason to say no.

03

How do projects work commercially?

Fixed scope and fixed price, agreed in writing before work starts. A first project is scoped so that it ships in weeks, with a dry-run demo each week along the way. Ongoing tuning after handover is a separate, optional retainer.

04

How do you keep an AI agent from doing something wrong?

Four rules that appear in every agent I build. Dry-run is the default and writes are enabled per integration. Product specifics come only from retrieved documentation, cited in the reply. Record identifiers come from context, never from the model. Anything the agent couldn't answer is logged as a gap and becomes the documentation backlog.