# Rishabh Singh (Rishabh K. Singh): the full record > AI engineer and co-founder shipping production software since 2019, starting at IBM Security. Co-founder and CEO of NestEazy, whose rental platform is live on the web, iOS and Android with six AI features in production. NestEazy has also built a multi-agent property OS, a proactive platform that helps landlords and property managers run their rentals while a person approves every decision that matters. It is in a pilot, not yet generally available. Updated 2026-10-30. Generated from the same record the portfolio's agents answer from (https://rishabhsingh.ca). Not to be confused with other people named Rishabh Singh: this is the AI engineer in Fredericton, New Brunswick, Canada. ## At a glance - Lead title: AI engineer. Current titles: Co-founder & CEO, NestEazy; Senior Software Developer, The Cabin Depot. - Location: Fredericton, New Brunswick, Canada. - Looking for: AI Engineer, Applied AI, forward deployed, founding engineer, agent infrastructure and AI security roles, at SDE II level or above. - Résumé versions: Applied AI Engineer, Backend Engineer, Agent Infrastructure Engineer, AI Security Engineer, Forward Deployed Engineer, Product Engineer, Full-Stack Engineer, Cloud Infrastructure Engineer, Software Engineer, Founding Engineer. Also searched as: LLM Engineer, Generative AI Engineer, AI Agent Engineer. - Relocation and remote: Open to relocating anywhere in Canada or the United States, and to remote or hybrid roles. - Work authorization: Lives in Canada; no sponsorship needed in Canada. A US-based role would need visa sponsorship. - Notice: Four weeks for a remote role, six to eight weeks for a role in Canada including a move, and ten weeks or more for a role in the United States. - Clearance: No government security clearance; willing to apply for one. - Education: Bachelor of Computer Science, University of New Brunswick (2017–2022); Diploma in Technology Management and Entrepreneurship; Dean's List; Edwin Jacob Special University Scholarship. Part-time data science program at Scaler Academy (2023–2025). - Links: https://rishabhsingh.ca/resume/ · https://ca.linkedin.com/in/rsingh11unb · https://github.com/nesteazy-rishabh · rishabhkumarsingh030@gmail.com ## Experience ### Co-founder & CEO, NestEazy Inc. Jan 2026 – Present · Fredericton, NB · alongside a full-time role - Built NestEazy's rental platform for the web, iOS and Android in six months on a two-person team. It launched in July 2026 and reached hundreds of users in its first city. - Shipped six AI features to production, including two embedding-based matching algorithms, document fraud checks and Ask Nesty, a retrieval-grounded (RAG) rental-law assistant on the website, used by over 1,000 people. - Lead NestEazy's multi-agent property OS (in a pilot), whose AI agents run leasing for landlords from first enquiry to move-in and hold anything consequential for their approval. - Co-designed the property OS so the model never acts alone. The model drafts and labels with no tools, code writes every automatic message, and choosing a tenant always needs human-in-the-loop (HITL) approval. - Run NestEazy's live web, iOS and Android platform with my co-founder and CTO, from deployment and monitoring to cost control. - Built a compliance engine for Canadian rental law across 13 jurisdictions, with sourced rules that return lawful notice dates and fill 17 official forms. - Built a video agent that turns listing photos into narrated, captioned tours, within per-run cost limits. - Won places in AWS Activate, the Google for Startups Cloud Program and Anthropic's startup program. - Build partnerships with landlords, property managers, local organizations, and cloud and AI providers, and own the roadmap and the company's corporate and tax administration. ### Senior Software Developer, The Cabin Depot Aug 2024 – Present · Fredericton, NB - Launched a B2B channel for 50+ installer partners across Canada and the US, with over half a million dollars in 2026 revenue, and built its partner portal (TypeScript, Express, Shopify GraphQL). - Built and operate a production AI-assisted email-security system that protects about 100 staff and partner inboxes from phishing, payment fraud and impersonation. - Run the technology, including about 8 Shopify stores and their integrations, for an off-grid living retailer selling across Canada and the US. Work with sales, management and outside vendors. - Resolved cross-system production issues between Shopify and the ERP, tracing inventory, pricing and order sync through SKUs, product classes and variant IDs. - Led a 5-person team through a from-scratch SYSPRO ERP rollout, from hosting, servers, VPN networking and user permissions to the database, Shopify data migration and integrations. Connected it to Power BI with least-privilege access. - Built a competitive intelligence platform (Python, Playwright, Flask, Shopify GraphQL) that tracks competitors' pricing, product and branding changes for about 4,000 SKUs across about 50 brands. It checks that matched models are the same product and maps substitutes for backordered items. ### Consultant, PLATO Testing Oct 2023 – Mar 2024 · Fredericton, NB - Automated an insurance client's Guidewire workflows as CenterTest (Java) suites in Azure CI/CD pipelines, working with the client's team in its own tools and triaging failures in Report Portal and Jira. ### DevOps Developer, QRadar on Cloud, IBM Canada Jan 2023 – Sep 2023 · Fredericton, NB - Deployed services for QRadar on Cloud, IBM's SaaS SIEM for enterprise customers, to Kubernetes (K8s) with Helm, and built Prometheus and Instana monitoring. - Handled incident response for production issues in deployments, networking and cluster resources. ### Junior Software Developer & Business Analyst, Geoffrey Stephens Engineered IT May 2021 – Nov 2022 · Fredericton, NB - Added computer-vision fraud detection to the Ableassess anti-fraud platform (ASP.NET MVC, Azure Computer Vision) and prototyped it on AWS Rekognition. - Built features for New Brunswick's COVID-19 vaccine booking system (Google Maps API) and its driver's licence test booking system (.NET, SQL). ### Security Software Developer (Co-op, then part-time), IBM Canada Jan 2019 – Dec 2020 · Fredericton, NB - Built log viewer and configuration tools (React, REST) for WinCollect, QRadar's Windows log agent, plus Python REST APIs for internal security tools, all following Security and Privacy by Design. - Brought SonarQube static analysis into the team's Jenkins pipelines to catch bugs and bad coding practices before production. Wrote Groovy, Geb/Spock and PowerShell test automation. ## Projects ### Agentic portfolio https://rishabhsingh.ca An MCP server (2026-07-28 spec), an A2A agent and an LLM tool-use loop behind a permission gateway. Its published eval passes 94.9% of 177 graded answers (75.6% on 15 held-out questions) at a p95 of 4.6 s and $0.006 an answer, and none of 88 red-team attacks changed what it did. ### CityCue https://citycue.ca A live, bilingual city feed for transit, roads, weather, outages and events across 13 Canadian cities, built with AI coding agents on dozens of public data feeds. Every update is sourced and timestamped. ### Garm https://github.com/nesteazy-rishabh/garm An open benchmark of six prompt-injection defence designs plus a staff-callback control, for agents that run as scheduled jobs. Per-field authority labels cut approvals on routine work to zero (against 102 to 163 per 100 tasks, partly by gating fewer actions). The callback stopped a handbook-following attack in 8 of 8 episodes. Run on two Claude models; 11 models from 7 vendors cross-checked the write-up. Code and run data are public. ## Skills - AI & agents: LLM APIs (Anthropic, OpenAI, Google), multi-agent orchestration, tool use, MCP (Model Context Protocol), RAG (retrieval-augmented generation), embeddings and vector search, PyTorch and ONNX inference, learning to rank, LLM evals, prompt and context engineering, guardrails, LangGraph - Languages: Python, TypeScript, JavaScript, SQL, Java, C#, Go - Infrastructure: CDN and secure document storage, high availability, AWS (EC2, S3, CloudFront, OpenSearch, CloudWatch, Bedrock), Google Cloud (GCP, Vertex AI), Azure, Docker, Kubernetes (K8s), Helm, nginx, Git, CI/CD (Jenkins, Azure Pipelines), observability (Prometheus, Instana, Sentry) - Security: prompt-injection defence, red teaming, least privilege, IAM, application security (SAST), SIEM - Product & data: React, Next.js, React Native (Expo), FastAPI, Node.js (Express), GraphQL, PostgreSQL (Postgres), SQL Server, SQLAlchemy and Alembic, Redis, Celery, pytest, Shopify, Supabase, Firebase (Cloud Messaging), Power BI ## Model card (skills, written like a model's) - AI features (In production): NestEazy: six, including document fraud checks, two ML matching algorithms, a reference agent and the Ask Nesty assistant. The Cabin Depot: an AI-assisted email-security system. This site's agent, graded by a published eval. - Agents & orchestration (Built): NestEazy's multi-agent property OS, which runs leasing from enquiry to move-in with the landlord approving every consequential action (in a pilot) - Vision, video & voice (Built): Vision checks on documents in production; a photos-to-video tour agent; a voice leasing prototype; computer-vision fraud detection earlier - Tool use & MCP (In production): This site's Streamable-HTTP MCP server, A2A agent, tool-use loop and Agent Skill; integrations with messaging, payments and e-commerce APIs (Twilio, Stripe, Shopify GraphQL) - Prompt & context engineering (In production): Ask Nesty's retrieval-grounded answers with citations and the listing agent's strict schema at NestEazy; this site's agent prompt (chapter 1), tuned against its own eval - Guardrails & AI security (In production): An email-security system protecting about 100 staff and partner inboxes; IBM Security background; this site's pre-model checks and a red team of 88 attacks (chapter 6) - Cost & reliability (Built): Budgets and safe fallbacks in the LLM layer of NestEazy's multi-agent property OS (in a pilot); prompt caching, per-turn token tracking and rate limits on this site - Evals & testing (Built): NestEazy's eval scripts (cost, p95 latency, injection cases, blind human grading); this site's published agent eval and a red-team suite run against the live agent; test automation at PLATO and IBM; Garm, an open benchmark with its code and run data public - Retrieval & ranking (In production): RAG over human-reviewed law for Ask Nesty; two embedding-based matching algorithms behind NestScore; BM25 ranking on this site - Production infrastructure (In production): Operates NestEazy's live platform; Kubernetes, Helm and Prometheus at IBM; this site on Vercel - Data & databases (In production): Databases in production at NestEazy; the database under an ERP rollout; Redis and SQLite; Power BI reporting - Shipping products (In production): NestEazy on iOS, Android and the web; CityCue; a B2B portal for a channel with over half a million dollars in 2026 revenue Limitations: - Garm (October 2026) is his first public benchmark; it is not peer-reviewed. - Works at the application, agent and infrastructure layers; has not pre-trained or fine-tuned a foundation model. - NestEazy's multi-agent property OS is in a pilot. It isn't generally available yet. ## Quick answers ### Who is Rishabh Singh? Rishabh Singh is an AI engineer in Fredericton, New Brunswick, Canada. AI engineer and co-founder shipping production software since 2019, starting at IBM Security. Co-founder and CEO of NestEazy, whose rental platform is live on the web, iOS and Android with six AI features in production. NestEazy has also built a multi-agent property OS, a proactive platform that helps landlords and property managers run their rentals while a person approves every decision that matters. It is in a pilot, not yet generally available. ### What roles is Rishabh Singh looking for? AI Engineer, Applied AI, forward deployed, founding engineer, agent infrastructure and AI security roles, at SDE II level or above. On job boards that's an AI engineer, applied AI engineer, LLM or generative AI engineer, or AI agent engineer, and the résumé has a version for 10 titles. Open to relocating anywhere in Canada or the United States, and to remote or hybrid roles. Anywhere includes Toronto, Vancouver, Montréal, San Francisco and New York. ### Where can he work, and does he need sponsorship? Lives in Canada; no sponsorship needed in Canada. A US-based role would need visa sponsorship. Notice: four weeks for a remote role, six to eight weeks for a role in Canada including a move, and ten weeks or more for a role in the United States. ### What has he built with AI agents? At NestEazy he co-founded a rental platform with six AI features in production and co-designed a multi-agent property OS, now in a pilot, in which the model has no tools and a person approves anything consequential. This site is an agent too: an MCP server, an A2A agent and a chat agent whose published eval passes 94.9% of 177 graded answers, and none of 88 red-team attacks changed what it did. He also published Garm, an open benchmark for prompt-injection defences in scheduled agents. ### How can an AI agent check his fit for a role? Connect any MCP client to https://rishabhsingh.ca/api/mcp and call match_job with a job posting, or tailor_resume for a résumé ranked for it. The A2A agent card is at https://rishabhsingh.ca/.well-known/agent-card.json, and https://rishabhsingh.ca/llms-full.txt holds the full record as text. ## The portfolio agent's published numbers pass@1 94.9% on 177 graded answers (59 questions, 3 runs each), 75.6% on 15 held-out questions, p95 4.6 s, $0.006 an answer; red team: 0 of 88 attacks changed what it did (run 2026-10-28).