Skip to content
Muhammad Ekram — home

00 / Abstract

Muhammad Ekram

AI Engineer building agentic systems, RAG pipelines, and the evals that keep them honest.

Table 1Headline results from project evaluations
injection attacks contained

injection attacks contained: 2/2

src
Notes Agent
n
2 attacks — an injected web page, a poisoned shared note
method
model scripted to obey the injection; pass = notes intact, nothing written outside notes/, attempt logged
hit@5

hit@5: from 75% to 83.3%

src
Documentation Q&A (RAG)
n
36 questions · 141 docs
method
hit = ≥1 relevant chunk (right file, and required text if set) in the top 5; baseline vs tuned config

01 / Projects

P01

Notes Agent

A tool-calling agent that manages a folder of Markdown notes, served as a CLI, an MCP server and an eval suite from one tool registry. Its prompt-injection defences live in code — a path allowlist and a human-approval gate — so they hold even when the model is fooled.

  • Python
  • MCP
  • Pydantic
  • FastAPI
  • pytest
  • Docker
  • GitHub Actions
Read the write-up: Notes Agent

injection attacks contained

injection attacks contained: 2/2

n = 2 attacks — an injected web page, a poisoned shared note

P02

Documentation Q&A (RAG)

Question answering over the FastAPI docs, grounded in the source with a citation on every claim, running fully offline. A 36-question golden set took hit@5 from 75% to 83.3% — and showed that two of three "obvious" retrieval upgrades were neutral or harmful.

  • Python
  • LangChain
  • FAISS
  • BM25
  • bge-small-en-v1.5
  • Qwen2.5-3B
  • llama.cpp
  • FastAPI
  • Docker
Read the write-up: Documentation Q&A (RAG)

hit@5

hit@5: from 75% to 83.3%

n = 36 questions · 141 docs

02 / Experience

  1. Remote, Toronto

    Applied AI Engineer Intern · Vosyn Inc.

    Agentic pipeline components on GCP.

  2. Software Engineer · Harlen Automation

    FastAPI backends and Python IoT control software.

Education

B.Tech CSE (AI concentration), 2022

Table 2Certifications (26), grouped by area
TitleIssued
Evaluation & safety
Evaluating AI AgentsArize AI—
Evaluating and Debugging Generative AIWeights & Biases—
Safe and reliable AI via guardrailsGuardrails AI—
Agents & LLM applications
Agentic AIDeepLearning.AI—
Multi AI Agent SystemsCrewAI—
Practical Multi AI Agents and Advanced Use CasesCrewAI—
MCP: Build Rich-Context AI AppsAnthropic—
Semantic Caching for AI AgentsRedis—
Generative AI with Large Language ModelsAmazon Web Services—
Gemini for Developers SpecializationGoogle DeepMindAug 2025
Retrieval & data
Advanced Retrieval for AI with ChromaChroma—
Building Multimodal Data PipelinesSnowflake—
Show 14 moreML foundations, Professional certificates (AI), Software engineering & DevOps, Other
TitleIssued
ML foundations
Machine Learning SpecializationStanford University—
Deep Learning SpecializationDeepLearning.AI—
Machine Learning in ProductionDeepLearning.AI—
Professional certificates (AI)
AI Engineering Professional CertificateIBMJul 2026
Generative AI Engineering Professional CertificateIBMMar 2026
AI Developer Professional CertificateIBMFeb 2025
Artificial Intelligence (A.I.) Professional CertificateGoogleMar 2024
AI Essentials SpecializationGoogleSep 2025
Software engineering & DevOps
DevOps and Software Engineering Professional CertificateIBMJun 2026
Applied DevOps Engineering Professional CertificateIBMFeb 2026
Applied Software Engineering Fundamentals SpecializationIBMMay 2026
Python 3 Programming SpecializationUniversity of MichiganNov 2025
Other
Cybersecurity Professional CertificateGoogleJun 2025
People Management Essentials SpecializationGoogleJan 2023

03 / Skills

Agents
  • LangGraph
  • MCP
  • Multi-step tool calling
Retrieval
  • Hybrid retrieval
  • Re-ranking
  • FAISS
  • pgvector
  • Pinecone
Evaluation
  • RAGAS
  • LangSmith
  • promptfoo
ML
  • PyTorch
  • LoRA fine-tuning
Infra
  • Python
  • FastAPI
  • Docker
  • AWS (Bedrock, EC2)
  • GCP (Vertex AI)

04 / Contact

Email is the fastest way to reach me.

hello@ekram.io