Skip to content
BackNov 7
Muhammad Tahir  .

Muhammad Tahir .

AI Agent Developer | Claude, MCP & LLM Integrations | Java Backend

I build AI agents and assistants that work in production: connected to your real systems, secured properly, and backed by solid backend engineering. Most AI agent projects stall after the demo. The agent can't reliably reach your data, breaks on edge cases, or opens security holes: leaked API keys, prompt injection, agents with more access than they need. I build agents the way I build backends: tested, secure, and maintainable. WHAT I BUILD - AI agents and assistants with Claude and other LLM APIs: tool calling, multi-step workflows, human approval steps - MCP (Model Context Protocol) servers that connect Claude to your APIs, databases and internal tools - Document Q&A and knowledge assistants (RAG) over your PDFs, docs and data - LLM features inside existing Java / Spring Boot applications (Spring AI, LangChain4j) - Workflow automation: agents that take repetitive tasks off your team - Security for AI systems: access control, OAuth / Keycloak, secrets handling, prompt-injection safeguards WHY WORK WITH ME - $20K+ earned and 2,000+ hours on Upwork building Java and Spring Boot backends, including long-term work on healthcare and enterprise platforms - Research internship at Ericsson: secure communication between microservices using mTLS, and certificate lifecycle management - M.Sc. in Computer Engineering (Software Platforms & Cybersecurity), University of Genoa, 2026. Thesis on confidential computing for trusted data ingestion - I work with Claude every day, so I know where LLMs are reliable and where they need guardrails ALSO AVAILABLE Java & Spring Boot backend development, REST APIs, microservices, Spring Security, PostgreSQL / MySQL / Oracle, Docker, AWS. HOW I WORK I start with a small working prototype on your real use case, so you see the agent working before committing to a bigger build. Then I harden it: error handling, logging, access control, and documentation your team can maintain. Based in Italy (CET), with working-hour overlap for European and US East Coast clients. Send me what you want to automate or use the Invite button, and I'll reply with how I'd approach it.

🇮🇹Genoa, Italy$25/hr100% JSS4.87 (16)Since Jan 2021

Earnings and projects

MRR
$58

Last 6 months ÷ 6

World#148,023 / 412k
🇮🇹Country#995 / 3,095
Revenue, last 30 days
$0

Last 6 months: $346

Total earnings
$23k

Billed in all: $24k

World#130,121 / 412k
🇮🇹Country#766 / 3,095
Per project
$998

$87 in the last 6 months

World#124,086 / 412k
🇮🇹Country#720 / 3,095
Projects, last 6 months
4
75% fixed25% hourly
World#51,813 / 412k
🇮🇹Country#359 / 3,095
Projects in all
23

Across 29 contracts

76% fixed24% hourly
World#120,400 / 412k
🇮🇹Country#904 / 3,095

Earnings, last 30 days

$24.1kNothing earned in the last 30 daysBilled in all, before Upwork's fee · drag across the chart to see what was earned in any stretch
$23,750$24,000$24,250$24,500No dataOct 8Oct 23Nov 7

Read once a week before Oct 15, 2026, daily since. Dotted: longer than usual between two readings. Hatched: over two weeks without one.

Skill rankings: Top 20% in Spring Security

Ranked by MRR among everyone on Upwork with the skill.

  1. #50Spring Security248 freelancers · 🇮🇹 #1Top 20%
  2. #78Application Security251 freelancers · 🇮🇹 #1Top 31%
  3. #132LLM Agent Development242 freelancers · 🇮🇹 #1Top 55%
  4. #137OAuth381 freelancers · 🇮🇹 #1Top 36%
  5. #221Model Context Protocol (MCP)413 freelancers · 🇮🇹 #1Top 54%
  6. #499Spring Boot2,150 freelancers · 🇮🇹 #1Top 23%
  7. #571Microservice2,217 freelancers · 🇮🇹 #1Top 26%
  8. #809AI Model Integration1,618 freelancers · 🇮🇹 #3Top 50%