You will get a Telegram or email AI bot with Claude or GPT and tool integrations
Project details
I build AI bots and agents that do real work - not just chat. Telegram bots that
post scheduled reports or answer questions, email assistants that read, summarise
and draft replies, or multi-tool agents that connect to your data and take actions.
What I build:
• Telegram bots: scheduled reports, alerts, Q&A, command handling
• Email assistants: inbox digests, auto-summaries, drafted replies you approve
• AI agents with tool use - calling APIs, reading/writing your data, taking actions
• Custom GPTs and Claude assistants with a precise system-prompt architecture
• MCP (Model Context Protocol) setups connecting models to your tools and data
I don't ship a prompt and disappear. I design the instruction layer, add guardrails,
and iterate the bot against your real cases so it holds up in production. You get a
working system you own and can edit.
Stack: Anthropic Claude API - OpenAI - Telegram Bot API - n8n - MCP - Python / JS
post scheduled reports or answer questions, email assistants that read, summarise
and draft replies, or multi-tool agents that connect to your data and take actions.
What I build:
• Telegram bots: scheduled reports, alerts, Q&A, command handling
• Email assistants: inbox digests, auto-summaries, drafted replies you approve
• AI agents with tool use - calling APIs, reading/writing your data, taking actions
• Custom GPTs and Claude assistants with a precise system-prompt architecture
• MCP (Model Context Protocol) setups connecting models to your tools and data
I don't ship a prompt and disappear. I design the instruction layer, add guardrails,
and iterate the bot against your real cases so it holds up in production. You get a
working system you own and can edit.
Stack: Anthropic Claude API - OpenAI - Telegram Bot API - n8n - MCP - Python / JS
Programming Languages
JavaScript, Python, TypeScriptCoding Expertise
Performance OptimizationWhat's included
| Service Tiers |
Starter
$40
|
Standard
$135
|
Advanced
$370
|
|---|---|---|---|
| Delivery Time | 3 days | 5 days | 8 days |
Number of Revisions | 1 | 1 | 2 |
Install Script | - | - | |
Test Script | - | ||
Task Automation |
Optional add-ons
You can add these on the next page.
Fast Delivery
+$35
Additional Revision
+$20
Extra channel
+$45
30-day support
+$60
Knowledge base / RAG
+$80Frequently asked questions
2 reviews
(1)
(1)
(0)
(0)
(0)
This project doesn't have any reviews.
JP
Jim P.
Jun 1, 2026
Meta Ads Creative (Premium Ecommerce – Video + Image Variations)
Very clear communicator and high standard of creativity. A learning for us both is that on AI projects the importance of clarifying detail before briefing and also work submission is key. However this was largely sorted with a very good result. Daniel is open, professional, creative and good to work with.
AO
Ayli O.
Apr 28, 2026
Senior AI Content Engineer / AI Influencer Builder (TikTok & Instagram Growth)
Great. Thank you.
About Daniel
AI Engineer | MCP Servers, RAG + Citations, LangGraph Agents, Evals
100%
Job Success
Eeklo, Belgium - 5:10 am local time
✔ agentic-rag-mcp (open source): a multi-agent RAG server over MCP where every answer cites its exact source chunk or the agent refuses. The eval suite — groundedness, citations, refusal correctness — runs in CI, so a regression fails the build.
✔ mcp-vitals (open source): a CLI that grades any MCP server A–F for reliability and agent-usability. I graded the official reference servers — two got an A, one got an F. Live report published.
✔ groundcheck (open source): a 1.5B groundedness-judge I fine-tuned (QLoRA) that agreed with a frontier judge 100% of the time at $0 per call — shipped only because its own evals beat the baseline.
✔ drumvia (live in production): a SaaS marketplace built solo end to end — auth, Stripe payments, KYC, real-time bidding, row-level security. Next.js, React 19, Supabase.
WHAT I DELIVER
• MCP servers — your API exposed to Claude/ChatGPT/agents: strict schemas, auth, remote deployment, observability, evals. Fixed scope.
• RAG systems with real citations — hybrid retrieval (pgvector) + reranking; grounded answers, measured quality.
• AI agents & automation — LangGraph pipelines, tool-calling, deterministic gates, human-in-the-loop; n8n + Claude workflows.
• AI reliability & evals — golden datasets, LLM-as-judge, eval harnesses wired into your CI.
HOW I WORK
Evaluation-first. I fix problems at the right layer — tool, gate, or prompt — then lock each fix with a regression test. Clear milestones, funded escrow, async updates. The boring reliability stuff — retries, idempotency, graceful failure — is what actually keeps a system alive.
Tell me what your AI is supposed to do, and I'll tell you how to prove it does it.
Steps for completing your project
After purchasing the project, send requirements so Daniel can start the project.
Delivery time starts when Daniel receives requirements from you.
Daniel works on your project following the steps below.
Revisions may occur after the delivery date.
Discovery & design
Define what the bot does, channel, data/tools - confirm with you.
Build & guardrails
Build bot, system prompt, tool use; add guardrails; iterate on real cases.