Robota
Autonomous job application pipeline. Five layers, three human gates. Discovers opportunities, researches companies, generates tailored documents — and never auto-submits.
Quality over quantity
Mass-apply tools send hundreds of generic applications. This system sends one — researched, tailored, and reviewed by a human before submission.
Manual applications convert at 6.6–10%. Auto-apply tools report 0.1%. This pipeline automates the research and generation, not the judgment.
The Pipeline
Follow a single job through the pipeline
Click each stage to see what happens to a real listing as it moves from discovery to delivery.
Real Output
Founded: 2006 · Specialist data & analytics Revenue: £68.2M (2024, +12% YoY) Overall match: 0.82 Strengths: BI platform leadership, governance
Scanned: 847 · Dedup: 312 · Filter: 44 · Match: 6 ▸ Head of Data — Harnham (0.82) ▸ Dir. Analytics — NHS Digital (0.79)
# [Candidate] Head of Commercial Systems & MI Data leader with 8+ years in UK government, specialising in BI platform architecture…
Hiring Manager: PASS · 78% quantified · 92% keywords Technical: PASS · stack alignment strong AI Detection: PASS · 0 em-dashes
Architecture
Structured, queryable knowledge base. Ten JSON files — skills, experience, achievements, education, narrative — each with strict Zod schemas.
Multi-turn AI agent loop producing company intelligence briefings. 30–40 reasoning steps per run across 8+ MCP sources.
Three-phase pipeline: content selection, parallel generation, multi-perspective audit. Zero tools, single-turn, deterministic.
Daily cron pipeline scanning four boards. Normalises, deduplicates, pre-filters, then scores via LLM.
Event-driven Telegram bot tying layers together. Background workers, PID tracking, Cloudflare KV, React dashboard.
Tech Stack
Runs autonomously on a Proxmox LXC container. Discovery cron at 7am weekdays. Telegram bot in tmux. No cloud compute bills.