Hang on! Adding more layers.
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
Acrobot-A2C
August 7, 2019 – August 7, 2019
Advantage Actor Critic implementation over the Acrobot-v1 environment.
View Projectdqn-segmentation
May 30, 2019 – May 30, 2019
A Deep Q-Network based Reinforcement Learning agent that can perform the NLP task of segmenting a chunk of given texts based on their semantic meaning and context. The taken approach is to feed a window of sentences into the DQN and the agent has discrete actions to determine at what index does a new segment lie or is there no segment in the given window at all. Except for tweaks and optimizations, the approach does seem to work, how well though is a mystery yet unsolved. Work under progress.
View ProjectMalaria-Cell-Detector
March 9, 2019 – March 9, 2019
A Convolutional Neural Network model, trained on the Malaria Cell Image Dataset from kaggle.
View Projectdyna-maze
February 17, 2019 – February 17, 2019
An agent designed by the dyna-Q approach of simultaneous Direct RL and Model Learning as described in the book "An Introduction to Reinforcement Learning, Sutton et al. 2018" The environment used is almost exactly the same as described in the book with the exception of a small negative living reward of -0.01. Typically learns the best policy under 4 episodes.
View Projectdqn-pong
January 20, 2019 – January 20, 2019
A Deep-Q Neural Network implementation to solve the Gym's Atari Pong environment.
View Projectcartpole-cross-entropy
December 28, 2018 – December 28, 2018
PyTorch implementation of the OpenAI Gym's CartPole environment using filtered training batches that give a higher overall reward than specified percentile.
View Projectchrome-dino-dqn
December 26, 2018 – January 22, 2019
A Double-Deep-Q-Learning Network implementation to play the Chrome T-Rex game.
View ProjectJacks-Car-Rental
December 3, 2018 – December 3, 2018
The implementation of Jack's Car Rental problem in the "Reinforcement Learning (2018) Second Edition" book by Sutton and Barto, MIT Press publication. Used Tabular Q-Learning method to solve the task, using number of cars at each location as states, and moving the number of cars between the locations as actions. If unfamiliar with the problem,
View Project2D-AI-Agent
October 26, 2018 – October 27, 2018
Implementation of 2-Dimensional Reinforcement Learning agent for the Grid World Maze just using numpy.
View ProjectReal-Estate-Predictor
August 18, 2018 – January 31, 2019
A linear regression model for house prices.
View ProjectCultural Fit Analysis
The candidate's projects are exclusively personal and heavily focused on Reinforcement Learning and Machine Learning. While this demonstrates initiative and a deep interest in a specific domain, the lack of diverse project types (e.g., web development, backend services, data engineering) or team-based work makes it difficult to assess broader cultural fit or adaptability to varied technical challenges typical in a general Software Engineer role. The candidate's experience level is listed as 0, which aligns with the personal project focus.
Soft Skills & Operational Fit
Insufficient data to assess soft skills or operational fit. The candidate's profile primarily showcases technical project work without details on collaboration, problem-solving approaches, or communication in a team setting.