Freelancers using OpenAI in NairobiFreelancers using OpenAI in Nairobi
ProjectsPeople
Vincent Pasili's avatar
Vincent Pasili
proNairobi, Kenya
Turning your fintech and web3 ideas into shipped products
$50k+
Earned
45x
Hired
5.0
Rating
225
Followers
KITTL_EXPERT

expert

LUMMI_EXPERT

expert

+2
Turning your fintech and web3 ideas into shipped products
Cover image for AI-powered Environmental Analysis Platform
AI-powered Environmental Analysis Platform
3
18
Cover image for Fiberwave Website Redesign and SEO Migration
Fiberwave Website Redesign and SEO Migration
0
20
Cover image for Epic Surf Discovery Platform Design
Epic Surf Discovery Platform Design
0
46
Cover image for Magarifiles Car Archive Web Design
Magarifiles Car Archive Web Design
0
38
OpenAI(1)
Stephen Kisong'e's avatar
Stephen Kisong'e
Nairobi, Kenya
Cyber Security Analyst
Cyber Security Analyst
Cover image for Conducted an automated security assessment
Conducted an automated security assessment of a locally-hosted large language model, Dolphin 3.0 (Llama 3.1 8B), using Garak, NVIDIA's open-source LLM vulnerability scanner. The model was served through LM Studio's OpenAI-compatible REST API and tested from a Parrot OS environment running in VirtualBox. Configured Garak to target the model via a custom REST generator configuration, handling endpoint routing, Bearer token authentication, JSON response parsing, and extended timeouts for local inference. Executed nine distinct probe categories against the model, covering DAN jailbreaks, prompt injection, encoding-based bypass, malware generation, latent injection, real toxicity prompts, continuation attacks, grandma exploits, and package hallucination. The assessment identified that Dolphin 3.0, an uncensored model with no built-in safety guardrails, was vulnerable across the majority of probe categories. Findings were documented per probe, distinguishing where the model resisted versus complied, and translated into a clear, layered explanation of each attack type and its real-world security implications. Delivered the work as a complete, reproducible walkthrough, including environment setup, tool configuration, probe execution, and results interpretation, suitable for both technical practitioners and stakeholders newer to AI security.
0
132
Cover image for Adversarial Data Poisoning Attack on a Network Intrusion AI Mod…
Adversarial Data Poisoning Attack on a Network Intrusion AI Mod…
0
2
Cover image for I conducted an authorized AI
I conducted an authorized AI security assessment of Lily Cybersecurity 7B to evaluate how effectively a hardened system prompt could resist jailbreak and prompt injection attacks. Using Prompt Fuzzer, I ran 15 attack techniques against the model. The system prompt successfully blocked 8 attempts, including most roleplay and social engineering attacks. The successful bypasses mainly used translated or altered wording to disguise the intent of the request. This project demonstrates why system prompts should be supported by input validation, moderation, output filtering, and continuous AI red team testing.
1
124
Cover image for Designed and documented a secure
Designed and documented a secure environment for running Robin, an AI-powered dark web OSINT tool, inside a compartmentalized Qubes OS and Qubes-Whonix setup. The environment was structured to separate research activity, Tor-routed traffic, sensitive credentials, untrusted content, and final reporting. Robin was deployed through Docker inside a dedicated research qube, with traffic routed through sys-whonix and sensitive notes stored separately in an offline vault qube. Disposable qubes were incorporated for opening potentially unsafe links and files. The setup was built around security by compartmentalization rather than relying on a single tool for protection. Particular attention was given to Docker mount restrictions, credential hygiene, network-boundary verification, lawful research scope, and keeping raw research data isolated from personal or client environments. The final result was a repeatable AI-assisted OSINT workflow that supports faster search refinement, result filtering, investigation summarization, and structured reporting while maintaining stronger operational security and clearer separation between collection, analysis, and final output.
0
113
OpenAI(1)