Skip to content
View mmf-code's full-sized avatar
🎯
Focusing
🎯
Focusing

Highlights

  • Pro

Block or report mmf-code

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.

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
mmf-code/README.md

Atakan Yaman

mostly making things complicated until they're simple.

co-founder / cto at evre.ai building AI systems for high-stakes human interaction.


right now

EVRE

building infrastructure for conversations that are difficult to fake, difficult to practice, and expensive to get wrong.

real-time AI simulations where people can step into a situation, talk their way through it, and see what actually happened.

some things currently taking up unreasonable amounts of my brain:

  • real-time conversational agents
  • keeping characters coherent under pressure
  • turning messy context into believable scenarios
  • multi-agent orchestration and simulation state
  • voice-first interaction
  • evaluating what happens inside a conversation
  • making synthetic interactions feel unscripted
  • figuring out where models stop behaving like people

recent experiment:

we wanted to know whether an uncensored model actually makes a better role-play character.

79% vs 2% refusal looked dramatic.

then we put the models in the chair.

the interesting part changed.

uncensored models in the chair →


before the agents

I spent a lot of time making physical systems behave instead.

Multi-Agent Formation Control under Disturbances

ROS 2 swarm simulation under meteorological disturbances. decentralized control, PID, Type-2 fuzzy logic, formation dynamics, wind.

repo →

Formation Control under Dynamic Meteorology

decentralized fuzzy–PID formation control with dynamic atmospheric disturbances, PSO tuning and Type-2 FLS.

repo →

Autonomous Vehicle Simulation

kinematic bicycle dynamics, Kalman filtering and path tracking.

repo →

Nonlinear System Identification & Control

black-box system identification with regression, curve fitting, NARX neural networks, fuzzy logic and adaptive control.

repo →


other things I've built

BLE Environmental Sensor Node C firmware on Raspberry Pi 5. I²C sensing → circular buffers → on-device statistics → dynamic BLE advertising.

repo →

GPU-accelerated image segmentation image processing and watershed experiments.

repo →

Modern anemometer Raspberry Pi, sensors, meteorology and 3D printing.

repo →


somewhere in the stack

Python · C++ · C · TypeScript · MATLAB LLMs · Agents · Evals · ROS 2 · Linux Control Systems · Fuzzy Logic · Simulation · Voice

control engineering + meteorology background.

went from making machines hold formation in bad weather to making AI hold a conversation when things get uncomfortable.

apparently I like systems that become unpredictable when the environment changes.


x · linkedin · evre.ai · kaggle

Pinned Loading

  1. multi-agent-formation-control-under-disturbances multi-agent-formation-control-under-disturbances Public

    Python 1 1