runable execution agents and founder's office work. i want to help compound trust, not silent churn.
i spent hours on runable. the product, reviews, discussions.
the ambition is insane. general-purpose execution agents are the future.
may 2025 incorporation to 750k+ users, benchmark dominance, real traction.
you're at the stage where trust compounds or silent churn compounds.
that's why i wrote this.
users shouldn't guess what a run costs. predicted credit burn before execution, free retries on formatting failures, transparent breakdowns per step.
operational safety nets matter. formatting retries should be free. auto-repair on broken outputs. rollback when an agent run goes wrong.
ai support triage first. fast escalation paths. auto-refund on clear failures. issue clustering so the same bug stops hitting different users silently.
trust center, compliance docs, SOC2 roadmap, DPDP alignment, data residency options. enterprise buyers need this before they commit.
whatsapp is the wedge US players won't prioritize. india runs on it. runable should own that distribution layer before someone else does.
distribution infrastructure, not a feature bolt-on. trigger agents from whatsapp, get status updates back, run workflows where india already lives.
memory, continuity, long-running tasks. agents that remember context across sessions instead of starting from zero every time.
trxnd thesis applied inside runable: autonomous outbound, lead research, account intel, narrative generation, meeting prep pipelines.
ironclaw connection. when an automation breaks, detect it, retry with alternate paths, escalate only when human judgment is actually needed.
hindi, tamil, telugu, and more. not translation as an afterthought. execution that works in the languages people actually operate in.
session-based AI forgets workflows, preferences, GTM learnings.
build an org-wide persistent memory layer that learns company writing, conversions, brand voice, failed approaches.
it becomes the operational brain, not just a tool you prompt. real moat.
at saturn labs, hundreds of hours of robotics training data were being manually qc'd every day.
people were checking:
manually.
so i built and integrated an automated qc pipeline overnight. the next day the team stopped wasting engineering time on repetitive review work.
that's usually how i operate.
python, typescript, sql, flutter
without really separating them.
multi-agent Android automation system. Controls phones the way a human would: reads the accessibility tree, decides what to tap next, no APIs needed. Automates job applications, handles CAPTCHAs by handing off to the user, accepts commands via voice/PDF/Telegram, supports 15+ languages.
real-time voice-cloning video translator. Takes a voice sample, then during a live call transcribes, translates (Gemini), and speaks back in your cloned voice via Qwen3-TTS. Sub-545ms end-to-end.
LLM inference engine built from scratch. Dynamic batching, KV-cache rewrite from O(N²) to O(1) per token, HTTP + gRPC, Prometheus/Grafana observability.
agentic RAG pipeline for causal extraction from conversational data. LangGraph orchestration, adaptive reranking, LLM judges. F1 = 0.94. Placed 5th at Inter-IIT Tech Meet 14.0.
AI mental wellness platform. Real-time conversational avatar via Gemini audio APIs, emotion-based music generation via Lyria, RAG-powered Persona Dashboard. Scaled to 1,000+ concurrent users.
live content distribution engines. Content routing, scheduling, channel output, reddit intelligence.
i like ambiguity, speed, pressure, undefined ownership.
founder's office is where i do my best work. no scoped tickets. no waiting for permission. just problems that need solving now.
give me a problem and 24 hours. i'll either have a working fix or a clear reason why it needs more.
that's the lane i want at runable.