Nishant L.
Claude & AI Automation Expert | AI Agent Developer | n8n, Make
Claude Expert | AI Agent Development | AI Automation | Claude Code | Claude Cowork | Claude Skills | MCP | Model Context Protocol | n8n | Make .com | Workflow Automation | Business Process Automation | API Integration I build AI systems that take a whole job off a team's plate and keep doing it — triaging the inbox, qualifying and following up leads, researching and publishing, booking and rescheduling, reconciling data, producing the weekly report — and hand back to a human at the points that actually need judgement. Most AI builds die in week three: a demo that works twice, breaks quietly, and everyone goes back to doing the work by hand. I build the version that's still running in month twelve — approval gates where a task is too critical to fully automate, error handling designed in from the start, and documentation so your team can maintain it without me. I hold 8 certifications from Anthropic, the company that makes Claude. I have not found another freelancer on Upwork who holds them. 𝗪𝗛𝗔𝗧 𝗜'𝗩𝗘 𝗔𝗖𝗧𝗨𝗔𝗟𝗟𝗬 𝗦𝗛𝗜𝗣𝗣𝗘𝗗 - 70+ automations in production. Roughly 1,000 manual hours saved per year for clients in the US, UK and EU. - An always-on Claude research and publishing agent that took a site from domain rating 0 to 28 with zero manually written articles. - An n8n outbound engine that booked 17 qualified meetings in 30 days for a B2B SaaS client. - A Make + HeyGen + ElevenLabs prototype I built seeded a client's AI-cloning product, which did $1M in sales in a single month. - Serenities AI — an agentic app platform I built solo, end to end, with Claude Code as the build agent. #12 on Product Hunt on launch day. - An AI video pipeline that cut production from ~5 hours of manual work to ~15 minutes of review. 𝗔𝗨𝗧𝗢𝗡𝗢𝗠𝗢𝗨𝗦 𝗖𝗟𝗔𝗨𝗗𝗘 𝗔𝗚𝗘𝗡𝗧𝗦 — 𝘁𝗵𝗲 𝘄𝗼𝗿𝗸 𝗜 𝘄𝗮𝗻𝘁 𝗺𝗼𝘀𝘁 Agents built on Claude, Claude Code and Claude Cowork that hold a job rather than answer a prompt. Scheduled runs, real browser interaction where an API won't do, tool use, memory and context management, retries and error handling. - Custom Claude Skills (SKILL.md) that trigger consistently and follow your process the same way every time - MCP servers (Model Context Protocol) connecting Claude to your CRM, database, calendar, inbox and internal APIs so it performs real actions, not summaries - Claude subagents and multi-agent architectures — a builder agent and a separate verifying agent, so nothing ships unchecked - Claude Cowork setup: workspaces, shared knowledge bases, permissions, scheduled tasks, team rollout - Approval gates and escalation rules, so a human is pulled in exactly where a task is too critical to fully automate — and nowhere else - Guardrails against prompt injection and permission sprawl, because an agent with inbox and CRM access is a security surface Already have an agent that half-works? I take those over too. 𝗧𝗛𝗘 𝗦𝗬𝗦𝗧𝗘𝗠𝗦 𝗔𝗥𝗢𝗨𝗡𝗗 𝗧𝗛𝗘 𝗔𝗚𝗘𝗡𝗧𝗦 - n8n and Make .com automation — Make Certified Level 5 — Expert. Lead routing, enrichment, follow-up, reporting, inbox triage, data entry, CRM hygiene. - AI voice agents — ElevenLabs, Vapi, Twilio, Deepgram. Answer calls 24/7, book and reschedule, quote from a live price book, push leads to CRM — hard rules enforced in the database, not in a prompt. - Full-stack AI apps — Next.js, TypeScript, Supabase, Postgres, Stripe, Vercel — built by directing Claude Code against a written spec, with approval gates and review at every stage. - GTM and outbound automation — Apollo, Clay, Apify, Smartlead, GoHighLevel, plus deliverability done properly (SPF, DKIM, DMARC). 𝗪𝗛𝗬 𝗖𝗟𝗜𝗘𝗡𝗧𝗦 𝗣𝗜𝗖𝗞 𝗠𝗘 - 8 Anthropic certifications, every one independently verifiable and listed in full below this overview - Make Certified Level 5 — Expert, Make's highest certification tier - 7 years in tech, 3 of them full-time on production AI. This isn't a 2025 pivot — I was automating businesses before the current wave. - Fully remote on US, UK and EU hours since 2017. Response inside 4 hours. Tell me what you're trying to automate and I'll tell you honestly whether it's a good fit before we start. If Claude isn't the right tool for it, I'll say so before you hire me. Keywords: Claude Expert · Claude Code · Claude Cowork · Claude Skills · SKILL.md · Anthropic · Claude Consultant · Claude Developer · Claude subagents · MCP · MCP Server · Model Context Protocol · AI Agent Development · Agentic AI · Autonomous Agents · Multi-Agent Systems · AI Automation Expert · AI Automation Specialist · Workflow Automation · Business Process Automation · CRM Automation · n8n Expert · Make .com · Integromat · OpenClaw · Prompt Engineering · RAG · AI Consultant · AI Implementation · Voice AI · AI Voice Agent · ElevenLabs · Vapi · Retell · Twilio · API Integration · Supabase · Airtable · GoHighLevel · Apify · Apollo · AI Chatbot
Earnings and projects
Last 6 months ÷ 6
Last 6 months: $1.4k
Billed in all: $19,429$19k
$205 in the last 6 months
Across 40 contracts
Earnings, last 30 days
Skill rankings: Top 22% in Data Scraping
Ranked by MRR among everyone on Upwork with the skill.
- #83AI-Generated Voice-Over206 freelancers · 🇳🇵 #2Top 40%🇳🇵#2 of 2
- #239ElevenLabs638 freelancers · 🇳🇵 #1Top 37%🇳🇵#1 of 1
- #801AI Consulting2,321 freelancers · 🇳🇵 #1Top 35%🇳🇵#1 of 4
- #883Prompt Engineering2,859 freelancers · 🇳🇵 #3Top 31%🇳🇵#3 of 4
- #926Airtable2,685 freelancers · 🇳🇵 #2Top 34%🇳🇵#2 of 5
- #949AI Implementation2,382 freelancers · 🇳🇵 #4Top 40%🇳🇵#4 of 10
- #1,283Business Process Automation3,508 freelancers · 🇳🇵 #3Top 37%🇳🇵#3 of 7
- #1,812Automated Workflow5,123 freelancers · 🇳🇵 #3Top 35%🇳🇵#3 of 5