Greg Hogg

Experience

• Founder and CEO · MLNOW Inc.

  Jan 2020 - Present

  Greg is the founder and CEO of Toronto-based MLNOW Inc, which owns Algomap.io. Algomap has over 100k free users and a paid program for busy engineers looking to crack FAANG. Through MLNOW Inc, Greg has contracted with some of the biggest tech companies in the world, including but not limited to NVIDIA, Cohere, HP, Gitlab, and Pecan AI. Greg is an expert in interview skills such as DSA, System Design, and AI / Data Analytics. Algomap shares public testimonials how the platform / bootcamp has helped them land roles at companies such as Google, Slack, Amazon, Meta, Capital One, and more.

• Deep Learning Engineer · Deep Breathe

  Sep 2021 - Jan 2022

  As a deep learning engineer with the medical startup Deepbreathe.ai, Greg wrote the data pipeline and neural network architecture for this published work in lung ultrasound imaging. Together, Greg and DeepBreathe created the first classifier to detect a rare and fatal lung condition known as Pneumothorax with high accuracy, which is one of many medical AI projects that DeepBreathe is getting recognized for. The company is actively involved in state-of-the-art grade medical research, even making advances in deploying their AI models into devices in the military to quickly diagnose and provide solutions for our soldiers.

• Machine Learning Engineer · Loblaw Companies Limited

  Jan 2021 - Apr 2021

  Loblaws is a 50 Billion Dollar grocery giant with over 2000 stores spread across Canada. As a Machine Learning Engineer, Greg assisted in building a custom neural network to forecast demand, helping the storekeepers know which products they would need to purchase and stock ahead of time.

• Data Scientist · Economical Insurance

  Jan 2020 - Aug 2020

  Economical is a top-10 largest property and casualty insurance company, headquartered in Waterloo, Canada since 1871. Greg worked here as a data scientist, and ultimately trained and deployed an insurance fraud detection model to automatically detect faulty claims, helping propel the company / industry's increasing move to AI.

• Software Developer · Ontario Ministry of the Attorney General

  May 2019 - Aug 2019

  The Ministry of the Attorney General in Ontario, Canada is responsible for government tasks such as managing the justice system, handling prosecutions, drafting laws, and offering legal advice to the government and public. Greg was a software engineer here who helped developed internal applications, tools, and reports.

• Software Developer · Canadian Food Inspection Agency

  May 2018 - Aug 2018

  As its name suggests, the Canadian Food Inspection Agency (CFIA) is a federal-level division of the Canadian Government, responsible for safeguarding food, animals, and plants, enhancing the health and well-being of Canada's people, environment and economy. Greg's first official role as a software developer here involved automating the creation of business-process diagrams (cooler than it sounds) and presenting findings to senior government officials.

Education

• University of Waterloo

  Bachelor's degree · Statistics and Computer Science · 2017 - 2022

Skills

Systems Design · Teaching · Python (Programming Language) · Machine Learning · Data Science · Programming · Data Engineering · Deep Learning · Community Engagement · Django · Web Development · Cascading Style Sheets (CSS) · Computer Vision · SQL · Data Structures · Software Development · Object-Oriented Programming (OOP) · Algorithms · Statistics · Data Analysis · Artificial Intelligence (AI) · C# · Java · C (Programming Language) · Databases · HTML · Firebase · Google Cloud Platform (GCP) · R (Programming Language) · Git · C++ · Public Speaking · Communication · Team Leadership · Algorithm Design · Predictive Analytics

Recent posts (12)

• SidecarOct 14, 2026

  Can you find a pair in one pass? 🤔 You have a sorted list: 1, 2, 4, 7, 11, 15. Find two numbers that add up to 9. Checking every pair works, but it gets slow fast. 6 numbers is 15 pairs. 1 million numbers is about 500 billion pairs. The trick is two pointers. Put one at each end of the sorted list and add them up: → Sum too big? Move the right pointer left → Sum too small? Move the left pointer right → Sum equals the target? Found it Every step throws away one number for good, so you finish in at most n steps instead of n × n. Swipe for: 🚂 The "meet in the middle" analogy (1869 railroad) 🧮 Why checking every pair is too slow 👈👉 Both pointers starting at the ends 🧭 The 3 rules for which way to move 📊 A 4-step walkthrough that lands on 2 + 7 = 9 🧠 Why you can safely drop a number without missing the answer 🎯 4 practice problems: Two Sum II, Valid Palindrome, 3Sum, Container With Most Water Save this for your next coding interview prep session 📌 Stop grinding random problems. Want the Coding Interview Roadmap used by 100k+ engineers, with theory and video walkthroughs for everything? Start at algomap.io/roadmap 👉 (link in bio) Which pattern should we break down next: sliding window or binary search? Tell me in the comments 👇 #twopointers #codinginterview #leetcode #datastructures #algorithms

• SidecarOct 14, 2026

  Hit a dead end? Undo and retry. 🔙 That's backtracking in one line, and it's behind a lot of FAANG interview questions. Swipe through to see it on the N-Queens puzzle: 👉 Choose a move, explore it, undo it if it fails 👉 Watch 4 queens hit a dead end, backtrack, and find the answer 👉 See why checking early (pruning) beats brute force: 16.7 million boards vs 2,056 placements for 8 queens 👉 Get the code template you can reuse Once you know this loop, you can use it on N-Queens, Sudoku Solver, Permutations, and Word Search. Stop grinding random problems. 🎯 Want the Coding Interview Roadmap used by 100k+ engineers, with theory and video walkthroughs for everything? Start at algomap.io/roadmap (link in bio) Save this for your next interview prep session 📌 and send it to a friend who's grinding LeetCode. #backtracking #codinginterview #leetcode #datastructures #algorithms

• SidecarOct 12, 2026

  Your boss asks: "What were our total sales from day 1 to day 3?" You add them up. Easy. Then they ask again. And again. For a million different ranges. That's where prefix sums come in. 🧠 Instead of re-adding every time: 1️⃣ Keep a running total as you go through the list 2️⃣ Write each total down once 3️⃣ Answer any range with ONE subtraction Sales: 2, 3, 7, 5 Totals so far: 2, 5, 12, 17 Days 1 to 3 → 17 − 2 = 15 ✅ Same idea as a car's odometer: miles for a trip = end reading − start reading. Why it matters in interviews: • Looping every query: up to 1 trillion steps • Prefix sum: about 2 million steps Prefix sums show up in Range Sum Query (LC 303), Subarray Sum Equals K (LC 560), Product of Array Except Self (LC 238) and Contiguous Array (LC 525). 💾 Save this for your next interview prep session 📩 Send it to a friend who's grinding LeetCode Want the Coding Interview Roadmap used by 100k+ engineers, with theory and video walkthroughs for everything? Start at algomap.io/roadmap 👉 #codinginterview #leetcode #datastructures #algorithms #softwareengineer

• SidecarOct 12, 2026

  Merge sort is a classic interview algorithm, and it gets easier once you see how it works: ✂️ Split the list in half, again and again, until every piece has 1 item (1 item is already sorted). 🔀 Merge the pieces back together. Compare the two front items and take the smaller one. 📈 Why it’s fast: halving takes about log n levels, and each level merges n items. That’s O(n log n). A basic n × n sort needs about 1 trillion steps on 1 million items. It’s the idea behind LeetCode 21, 148, 23 and 315, so it’s worth knowing cold. Swipe through for the diagram, a step-by-step merge, and what the code has to do. 👉 🗺️ Stop grinding random problems. Get the Coding Interview Roadmap used by 100k+ engineers, with theory and video walkthroughs for everything: algomap.io/roadmap (link in bio) 💬 Which algorithm should we break down next? Tell us below. #mergesort #datastructures #algorithms #codinginterview #leetcode

• SidecarOct 11, 2026

  Why do Kafka and SQS even exist? 👇 Say checkout calls the email service directly. If email is slow or down, checkout stalls, and one failure spreads to everything after it. The fix is to put a queue in the middle: → Checkout drops a message and moves on → Email reads it when it’s ready → If email is down, the messages wait safely Swipe for: 🍔 The lunch-rush analogy 📉 A traffic spike, second by second (150 messages dropped without a queue) 🧱 Kafka in one picture: topics, partitions, consumers ⚖️ SQS vs Kafka: a queue for tasks, a log for history 🔁 Why messages can arrive twice, and why your code must be safe to repeat (idempotent) 🎯 4 system design prompts where a queue fits Save this for your next system design interview 📌 Stop grinding random problems. Want the Coding Interview Roadmap used by 100k+ engineers, with theory and video walkthroughs for everything? Start at algomap.io/roadmap 👉 (link in bio) Which would you like next: Kafka partitions or rate limiting? Tell me in the comments 👇 #systemdesign #softwareengineering #kafka #awssqs #messagequeue

• SidecarOct 11, 2026

  Which server gets the request? 🤔 Your app has 3 servers and a million users. Something has to decide who handles each request. That's the load balancer. Two classic ways it decides: 🔄 Round Robin: take turns. A, B, C, A, B, C. Simple, no tracking. 📉 Least Connections: pick the server with the fewest active requests. The catch? If some requests are slow, round robin can pile them all on one server while the others sit idle. Least connections avoids that, but it has to track a live count per server. Swipe through for: ✅ The checkout-lane analogy ✅ A step-by-step example (R1 to R6) ✅ Both algorithms in code ✅ When to pick which ✅ The follow-up questions interviewers love: weighted round robin, sticky sessions, health checks, backup balancers Load balancing shows up in almost every system design interview. Be ready to explain WHY you picked one, not just name it. Want the full Coding Interview Roadmap used by 100k+ engineers, with theory and video walkthroughs for everything? 👉 algomap.io/roadmap (link in bio) Save this for your next interview prep session 📌 #systemdesign #loadbalancing #codinginterview #softwareengineering #computerscience

• SidecarOct 10, 2026

  Best streak of days? Find it in ONE pass 👇 Given a list of numbers, which stretch of neighbors adds up to the most? That's the Maximum Subarray problem, and Kadane's Algorithm solves it in a single loop. The trick: ➕ Add the next number to a running sum 🔄 If the sum drops below 0, reset it to 0 (a negative sum only drags down what comes next) 🏆 After every step, save the best sum you've seen On −2, 1, −3, 4, −1, 2, 1, −5, 4, the best block is 4 + (−1) + 2 + 1 = 6. Why it matters: ⚡ Checking every block takes about 167 million steps on 1,000 numbers ⚡ Kadane's takes 1,000 steps and just 2 variables Swipe through for the walkthrough, the code, and the problems that use this idea: Maximum Subarray, Max Circular Subarray, Max Product Subarray and Buy and Sell Stock. Which algorithm should we break down next? Tell us in the comments 👇 Stop grinding random problems. Get the Coding Interview Roadmap used by 100k+ engineers, with theory and video walkthroughs for everything: algomap.io/roadmap (link in bio) #kadanesalgorithm #maximumsubarray #leetcode #codinginterview #datastructures

• SidecarOct 10, 2026

  Your database is slow, full, or one crash away from taking down your app. Do you copy it or split it? 🤔 Replication = copies of ALL your data on many servers. → Fixes heavy reads and server crashes. Sharding = a different slice of your data on each server. → Fixes huge data and heavy writes. Real systems do both: split the data into shards, then keep a replica of every shard. Swipe to see: 📚 The encyclopedia analogy 🔀 How user_id % 3 picks a shard ⚠️ The catches: replica lag, hot shards, cross-shard JOINs 🎯 4 interview questions and which fix to pick Save this for your next system design interview 🔖 Want the full Coding Interview Roadmap used by 100k+ engineers, with theory and video walkthroughs for everything? Start at algomap.io/roadmap 👉 #systemdesign #databases #sharding #replication #softwareengineering

• SidecarOct 9, 2026

  Fewest stops isn't always cheapest. 🗺️ Dijkstra's algorithm finds the cheapest route from one town to every other town. Swipe to see it step by step: 📓 Keep a notebook of the cheapest cost found so far 🏙️ Always visit the cheapest unvisited town next 🔀 A cheaper detour replaces the old cost (D went from 14 to 9) 🔙 Backtrack to read the route It shows up in LeetCode 743, 1631, 778 and 1514. Watch out for negative costs, which break it. Save this for your next interview prep session. 👇 Want the full coding interview roadmap? Link in bio: algomap.io/roadmap #dijkstra #datastructures #algorithms #codinginterview #leetcode

• SidecarOct 9, 2026

  Are these two items connected? 🤔 Union-Find answers that in almost no time. Every group gets one leader, and two items are in the same group if they share a leader. That one idea solves a whole family of interview problems. Swipe to see: ➡️ How leaders and parents work ➡️ Find: climb to the top ➡️ Union: merge two groups ➡️ Why path compression makes find fast ➡️ 4 LeetCode problems that use it (547, 684, 721, 261) Save this for your next interview prep session 📌 Stop grinding random problems. Get the full Coding Interview Roadmap used by 100k+ engineers at algomap.io/roadmap #leetcode #datastructures #codinginterview #softwareengineer #computerscience

• SidecarOct 8, 2026

  Socks before shoes. Prerequisites before advanced courses. 🧦👟 Whenever tasks depend on each other, you need a valid order. That's Topological Sort, and it's hiding behind a lot of FAANG interview questions. Swipe to learn: ➡️ Tasks become arrows (a DAG) ➡️ In-degree: who's blocking you? ➡️ Kahn's algorithm in 4 steps ➡️ A full walkthrough on a real graph ➡️ Why a cycle breaks everything Next time you see "can you finish all the courses?" you'll know exactly what to do. 💡 Save this for your next interview prep session 📌 Send it to someone who's grinding LeetCode 👀 Want the full coding interview roadmap used by 100k+ engineers, with theory and video walkthroughs for everything? Start at algomap.io/roadmap 👉 (link in bio) #leetcode #datastructures #algorithms #topologicalsort #softwareengineer

• SidecarOct 8, 2026

  Interviewers love this one: "Does this linked list contain a cycle?" You could track every node you've visited, but that costs O(n) extra memory. The better answer is Floyd's Tortoise & Hare: 🐢 slow pointer moves 1 step 🐇 fast pointer moves 2 steps 🔁 if there's a loop, fast eventually catches slow 🏁 if there's no loop, fast hits null and you stop Inside the loop, fast closes the gap by exactly 1 node every step, so it can never jump over slow. The result is O(n) time and O(1) space in 8 lines of code. Swipe through for the step-by-step walkthrough, the code, and the LeetCode problems where this trick shows up (141, 142, 202, 287). 👉 💾 Save this for your next interview prep session. 💬 Comment "LOOP" if you want a follow-up on finding where the cycle starts (LeetCode 142). Stop grinding random problems. Follow a step-by-step roadmap to your FAANG offer. Start at algomap.io/roadmap (link in bio). #codinginterview #leetcode #datastructures #algorithms #linkedlist