Skip to content
View mahdims's full-sized avatar
57 followers · 9 following

Achievements

Achievement: Pair ExtraordinaireAchievement: Pull Sharkx2Achievement: YOLOAchievement: Starstruck

Achievements

Achievement: Pair ExtraordinaireAchievement: Pull Sharkx2Achievement: YOLOAchievement: Starstruck

Block or report mahdims

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
mahdims/README.md

Mahdi Mostajabdaveh

Principal OR/AI Scientist at Kinaxis · AI research and technical leadership

I build AI systems for reasoning, algorithm discovery, and optimization. At Kinaxis, I lead AI acceleration for a production solver and evaluation of coding agents. Previously, I led AI research for Huawei's OptVerse solver, shipped a routing engine used by 20+ enterprise customers, and developed OptvEvolve.

Our OptVerse-CityU team won first place in the 2026 CVRPLIB Best Known Solutions Challenge, with 51 new best-known solutions.

Research: agents and memory; algorithm discovery and learned search; language to formal models; optimization methods. Work includes MASPRM, SEDIMA, EvoCut, Latent Heuristic Search, COAgents, ORQA, NL4Opt, and AILS-Enhanced.

I connect AI proposals with reliable tests: feasibility, objectives, bounds, benchmarks, and held-out instances. My work has appeared at NeurIPS, AAAI, ACL, EJOR, and INFOR. I am lead inventor on multiple filed patents.

Ph.D. in operations research; postdoctoral fellow with Michel Gendreau at Polytechnique Montréal.

Website · Publications

Pinned Loading

  1. Branch-and-price- Branch-and-price- Public

    This is my implementation of a branch and price algorithm to solve the humanitarian aid distribution problem. This problem is a VRP with a specific objective function

    Python 61 13

  2. nl4opt/nl4opt-competition nl4opt/nl4opt-competition Public

    Natural Language for Optimization Modelling

    Shell 78 9

  3. COAgents COAgents Public

    Official code for "COAgents: Multi-Agent Framework to Learn and Navigate Routing Problems Search Space" (LION 2026). Three learned agents (node selection, move selection, jump) orchestrate local se…

    Python 3

  4. milad1378yz/EvoCut milad1378yz/EvoCut Public

    EvoCut: Automatic generation of acceleration cuts for integer programs via evolutionary search and LLMs.

    Python 11 2

  5. researchmate researchmate Public

    Stay current on high-quality work at the Operations Research × AI intersection. A research agent performs deep, multi-pass analysis of each paper, scores relevance, and curates what matters, helpin…

    Python 11 1

  6. EvoCode EvoCode Public

    LLM-guided evolutionary framework for evolving code components, plug in your own domain, builder, and evaluator.

    Python