
In Progress
Posted
Paid on delivery
I’m building a proof-of-concept where an autonomous agent can watch a virtual world through an SDK, reason about what it sees and then act just like a fully fledged non-player character. The environment itself is NPC-centric, so everything you create has to plug cleanly into the existing observation / action interface I already expose. What the agent must ultimately demonstrate is the full spectrum of believable behaviour: • NPC behaviours to cover: combat actions, rich social interactions and day-to-day resource management. • Within combat the agent needs strategic movement, solid weapon-handling logic and the ability to switch into defence tactics when the situation flips. I’ll provide the SDK (Python and C++ bindings) plus stubbed observation data. Your job is to wire up the decision-making layer—whether you favour behaviour trees, GOAP, deep reinforcement learning or a hybrid approach is up to you—as long as the final agent can: 1. Parse state frames coming from the SDK in real time. 2. Select an action that makes sense for the current goal. 3. Return that action through the SDK with <50 ms round-trip on my test hardware. 4. Learn or adapt over multiple episodes so behaviours don’t feel scripted. I’ll review the work in a headless test harness first, then inside the full 3-D client. Make sure the code is documented and that I can swap out reward functions, tweak difficulty or add new action primitives without refactoring the core logic. If this fits your skill set—Unity or Unreal integration experience is a plus—let’s talk about milestones and get an AI NPC thinking for itself.
Project ID: 40644905
62 proposals
Remote project
Active 1 min ago
Set your budget and timeframe
Get paid for your work
Outline your proposal
It's free to sign up and bid on jobs

Hello There! I’m Md Toriqul Islam, an experienced AI developer specializing in autonomous agents, reinforcement learning, Python/C++, real-time decision systems, and game AI. I’m excited to partner with you and can start immediately. I have rich experience with AI decision-making, behavior trees, GOAP, adaptive agents, real-time state processing, and SDK/API integrations. I am skilled in Python, C++, reinforcement learning, behavior trees, GOAP, game AI, and Unity/Unreal integration. I understand you need a decision-making layer that consumes real-time SDK observations, selects context-aware NPC actions, and returns them within a strict sub-50ms round trip. The agent should support strategic combat, weapon handling, defensive tactics, social interactions, and resource management while adapting across episodes and keeping rewards, difficulty, and action primitives configurable. I’m ready to start immediately and would be happy to review the SDK, observation/action interfaces, and headless test harness to define milestones. Looking forward to hearing from you. Best regards, Md Toriqul Islam
$100 AUD in 3 days
4.5
4.5
62 freelancers are bidding on average $254 AUD for this job

Hello Sir/MAM I am a Skilled Full Stack Developer. Having rich experience in Java , C++ , C , C# , Python , Eclipse , Sql , Mysql , .Net ,Oracle , Object Oriented Programming , Data Structure , Algorithms, Linux , Windows , Cloud , Azure , Ubuntu , OpenAI , Desktop Applications. Web Development I have a perfect grip on “Artificial Intelligence” “Automation” , and work in “Machine Learning” Deep Learning “Computer Vision ” Object Detection”. My track record as demonstrated in my 100% job completion and 5-star review rating showcases My ability to deliver exceptional results on time and with utmost quality I believe that my skill set makes me the ideal candidate for this project Please come on chat we will discuss more about this I will be waiting for your reply . Thanks and Best Regards
$140 AUD in 1 day
6.2
6.2

I can help you build the decision layer as a modular, real-time agent pipeline: parse SDK state frames, normalize them into a compact observation space, then run a goal-driven behavior selector. I’d use a hybrid approach—GOAP or utility-based goal arbitration for high-level decisions, behavior trees for combat/social/resource execution, and a lightweight learning layer on top so behavior adapts across episodes without feeling scripted. I’ll keep action primitives, reward functions, and goal weights configurable so you can tweak difficulty or add new behaviors without refactoring the core loop. For the <50 ms requirement, I’ll profile the hot path, preallocate buffers, and make sure the observation-to-action round-trip is tight. I’ll also handle the edge cases around switching between social goals, resource management, and defensive combat tactics based on incoming state—so the agent reacts to the situation, not a fixed sequence.
$2,300 AUD in 7 days
5.6
5.6

✅Full Experience in AI NPC Automation & Behavior Development ✅. ✳️I am very confident that complete your project perfectly. ✳️I can guarantee the quality of the job and deliver the result on time. I hope we will discuss in more detail via chat. Best regards!
$100 AUD in 2 days
5.9
5.9

Drawing from my extensive 5+ years of Unity Game Development experience, I possess the skills necessary to bring your project to life. I strongly believe that my expertise in Unity, Unreal Engine, Roblox and immersive AR/VR experiences will be instrumental in creating a believable autonomous agent for your virtual world. Having built various gameplay systems and mechanics, I understand the importance of delivering smooth and optimized performances while still sustaining visually appealing environments. Moreover, my focus on performance, scalability, quality, and strong gameplay system design is directly aligned with the task at hand. From parsing real-time state frames to creating AI that varies its behavior over time, rest assured that I can create an AI NPC with a certain level of depth so that it does not feel scripted. With my skill set and robust problem-solving abilities, I promise a highly optimized output without compromising on consistency and believability. Finally, I pride myself on reliable communication and on-time delivery. I understand your need for flexibility in the project structure- availability to swap out reward functions or add new action primitives, all while maintaining an efficient core logic- and ensure that all my codes are well-documented to facilitate any future need. If you're seeking smooth project management coupled with outstanding-optimized outcomes, let's connect!
$400 AUD in 20 days
5.8
5.8

As a team specializing in agentic AI and system integration, our expertise aligns perfectly with your NPC automation and behavior development project. We not only specialize in building AI systems from the ground up but also ensuring their seamless integration into your existing workflow. Our familiarity with Python, C++, Unity, Unreal and more of the tools you mentioned will help us create the exact environment you seek for your SDK. Notably, we are experienced in integrating AI into ERP workflows, reading real-time sensor data, and acting upon it accordingly - skills that are directly valuable to your project's goals. Our proficiency with decision-making layers such as behavior trees, GOAP, deep reinforcement learning offer a comprehensive approach when designing autonomous agents. We understand that merely delivering a functioning proof-of-concept is not enough; it needs to be flexible and easy to adapt without disrupting critical components. We commit to providing well-documented, modular code that allows for easy reward function changes or addition of new action primitives. With our budget-conscious mindset and a dedication to timely turnaround, we are ready to take on this challenge and bring your virtual world NPC's believable behaviors to life!
$140 AUD in 7 days
3.9
3.9

Hi, Creating NPCs that appear intelligent isn't just about choosing actions—it's about building a decision architecture that can observe, prioritize, adapt, and react in real time. That's the challenge that makes this project exciting. With 8+ years of game development experience, including AI systems, gameplay architecture, state machines, behavior-driven systems, and Unity-based NPC development, I can build a modular decision layer that integrates cleanly with your SDK. My approach would evaluate GOAP, Behavior Trees, Utility AI, or a hybrid model depending on the observation space and performance requirements, while maintaining sub-50ms decision latency. The resulting system will be extensible, allowing new goals, actions, reward models, and difficulty adjustments without major refactoring. I'd be interested in reviewing the SDK, action space, and NPC requirements to propose the most effective architecture and development roadmap. Regards, Muhammad Asad Ali
$150 AUD in 7 days
4.2
4.2

With a strong background in C++ programming and AI automation, I can expertly handle the intricate requirements of your proof-of-concept. I have extensive experience with combat gaming mechanics, social interactions, and resource management — all key elements you seek for your autonomous agent. My approach to decision-making is agile, where I employ a hybrid of techniques like behaviour trees, GOAP and deep reinforcement learning to create NPCs that are not only reactive, but capable of learning and adapting over multiple episodes. Hitting your desired <50 ms round-trip within the SDK will be effortless for me as a Full Stack Developer accustomed to building high-performing software. My prior work with RAG applications, AI agents and agentic backends guarantee that I can parse state frames in real-time, select actions aligned with the current goal, and return the output promptly. I’ll ensure the code is well-documented, offering flexibility for you to adjust reward functions or add new actions to the core logic with ease. Expertise in Unity or Unreal could be an advantage here, but even without that specific experience my previous work integrating various platforms (like WordPress with Salesforce) demonstrates my ability to adapt quickly and drive efficient results. Overall, my commitment is to deliver not just reliable code but an effective solution that makes your AI NPC feel organic and engaging. Let's get started!
$150 AUD in 5 days
3.9
3.9

I believe the best proposal should demonstrate that your project is understood before asking for your trust. Your ambitious project to develop an autonomous NPC that exhibits believable behavior in a dynamic virtual environment is both exciting and complex. The need for strategic combat actions, nuanced social interactions, and effective resource management calls for a sophisticated decision-making framework that seamlessly integrates with your existing SDK. I propose employing a hybrid approach that leverages behavior trees and deep reinforcement learning to ensure adaptable and evolving NPC behaviors. This will allow the agent to respond intelligently to real-time state frames while maintaining the ability to learn from past interactions, enhancing realism and depth. My previous work on a similar AI-driven NPC project showcased how behavior trees can effectively manage complex decision-making, resulting in engaging user experiences. You will benefit from well-documented code, ensuring easy modifications to reward functions or action primitives, thus maintaining flexibility as your project evolves. By implementing this solution, your NPC will not only function effectively within the given constraints but will also exhibit intelligent behavior that captivates users. Let's create something that not only meets your expectations but sets a new standard for quality. Regards Junaid
$100 AUD in 7 days
3.8
3.8

Hello There! I’m Ruhul Ajom, an experienced AI and full-stack developer with 10+ years of experience, and I can dive into your project immediately. I understand you need an autonomous AI NPC decision-making layer that can process real-time observations from your Python/C++ SDK, reason about combat, social interactions and resource management, and return appropriate actions with low latency. I have rich experience in Python, C++, AI/ML, reinforcement learning, API integration, automation, and real-time application development. I am skilled in designing modular AI behaviour systems using behaviour trees, GOAP, reinforcement learning, or hybrid approaches, with configurable rewards, difficulty settings, and extensible action primitives. I can integrate the decision layer with your existing SDK, optimize the observation-to-action pipeline for the required latency, support adaptive behaviour across episodes, and provide clean documentation and testing through the headless harness. I’m ready to start immediately and would be happy to review the SDK and observation/action interfaces. Looking forward to hearing from you. Best regards, Ruhul Ajom
$50 AUD in 2 days
4.8
4.8

⚠️ IF YOU'RE NOT HAPPY YOU DON’T PAY ⚠️ Hi, I'm excited to develop an autonomous agent that demonstrates believable behavior in your virtual world using the provided SDK. I'll utilize Python and C++ to create a decision-making layer, potentially leveraging behavior trees or deep reinforcement learning to achieve strategic movement, social interactions, and resource management. With my experience in developing adaptive AI systems, I'm confident in delivering a well-documented codebase that meets your requirements. I’d love to chat about your project! The worst that can happen is you walk away with a free consultation. Regards, Chris
$100 AUD in 5 days
4.0
4.0

This is a strong fit for an autonomous NPC decision layer that must read SDK state, choose actions in real time, and adapt over repeated episodes. I can structure the system around a clean observation-to-decision-to-action pipeline that stays modular, low-latency, and easy to extend. For the core architecture, I would build a hybrid agent that combines behavior logic for reliability with learning components for adaptation. That lets combat, social behavior, and resource management remain controllable while still evolving across episodes. The implementation can be organized so new action primitives, reward functions, or difficulty settings can be added without changing the decision core. I can also optimize the round-trip path to respect the <50 ms target by keeping state parsing, policy selection, and SDK action dispatch lightweight and well-isolated. Clear documentation will be included so the headless harness and full 3-D client can both use the same interface cleanly. If helpful, I can break this into milestones covering SDK integration, baseline NPC logic, combat tactics, adaptive learning, and test-harness validation before final client verification.
$250 AUD in 4 days
2.4
2.4

Hi, I can develop the AI NPC decision-making layer for your proof-of-concept, using the provided SDK observation/action interface to support combat, social interaction, resource management, and adaptive behavior. The best solution is to first review your SDK bindings, observation state format, action primitives, latency target, test harness, and desired behavior goals. Then I’ll design a modular behavior system using behavior trees, GOAP, or a hybrid approach so the agent can parse real-time state, select sensible actions, and return decisions within the required <50 ms loop. I’m comfortable with C++, Python, AI agent behavior, NPC logic, behavior trees, GOAP-style planning, reinforcement/adaptive behavior concepts, SDK integration, headless test harnesses, combat decision systems, reward tuning, and Unity/Unreal-style game AI workflows. Deliverables will include: * SDK observation parser * NPC decision-making layer * Combat behavior logic * Social interaction behavior * Resource management behavior * Strategic movement/defense logic * Configurable goals/reward settings * Action selection under latency target * Headless test harness support * Documented source code * Extension notes for new actions I’ll focus on a clean, modular AI behavior system that feels less scripted, can be tuned over episodes, and remains easy to extend with new goals, difficulty settings, and action primitives. Best regards Ankit
$100 AUD in 2 days
2.3
2.3

Hello, As a result of a detailed review of your project requirements, I fully understand that the key challenge is building an adaptive NPC decision layer that plugs directly into your existing observation/action SDK while keeping inference under the 50 ms round-trip target. I have experience with C++/Python AI systems, real-time decision logic, game-agent architecture, reinforcement learning, and Unity-oriented integrations, and I’m available to start right away. For this POC, I’d recommend a hybrid architecture: behaviour trees or GOAP for deterministic high-level goals, combined with a lightweight learned policy for tactical choices such as combat positioning, weapon selection, defence switching, and adaptive resource behaviour. This gives you believable adaptation without sacrificing latency or debuggability. I’ll structure the agent around modular state parsing, goal selection, action arbitration, reward configuration, and pluggable action primitives so you can change difficulty or add behaviours without rewriting the core. I’ll also profile the hot path to keep decision latency within your target. I have two quick questions. • Is the headless harness already able to run repeated episodes automatically? • Do the SDK bindings expose synchronous step calls, callbacks, or both? I would be glad to discuss milestones and can start immediately. Best regards, Carlos
$50 AUD in 7 days
1.0
1.0

Hello, I HAVE CREATED SIMILAR PROJECTS BEFORE AND I CAN SHOW YOU. I have carefully gone through your requirements and understand that you need an AI-driven NPC behavior system that can process real-time SDK observations, reason about the environment, and execute adaptive combat, social, and resource-management actions. I have 13+ years of experience and can build a modular decision-making layer using Python/C++, behavior trees, GOAP, reinforcement learning, or a hybrid architecture based on the SDK and performance requirements. I will ensure real-time state processing, strategic combat movement, weapon handling, defensive behavior, social interactions, resource management, and adaptive behavior across episodes while targeting the required <50 ms round-trip performance. The architecture will keep observation, decision-making, rewards, difficulty settings, and action primitives modular so you can extend the NPC system without major refactoring. I will also provide documented source code and support integration with your headless harness and 3-D environment. I WILL PROVIDE 2 YEAR FREE ONGOING SUPPORT AND COMPLETE SOURCE CODE. WE WILL WORK WITH AGILE METHODOLOGY AND WILL GIVE YOU ASSISTANCE FROM ZERO TO PUBLISHING ON STORES. I am available according to your convenient time zone and can successfully implement this project from start-to-finish. I eagerly await your positive response. Thanks, Christina
$140 AUD in 7 days
5.0
5.0

Hello, I understand that you require AI NPC automation. You need smart NPC behaviors and automated decision systems built for Unity. I will use clean C++ and CUDA scripts to handle fast NPC logic. With 5+ years of experience and a 5 star rating, I am confident that I can provide you with the best solution regarding your AI NPC development. I assure you with professional communication until your work is done. Regards, Syed Anas
$60 AUD in 1 day
0.0
0.0

As an experienced developer focusing on app development, I have extensive knowledge of C and C++, essential skills for successfully parsing state frames from your SDK in real-time. Moreover, my dedication to crafting high-performance, user-friendly applications makes me a perfect fit for your project as these are key elements for creating a believable NPC. In addition to my programming proficiency, my deep understanding of UI/UX design ensures that the final product will be aesthetically pleasing and seamlessly integrate with your existing system. This includes providing adaptable code that allows you to tweak difficulties or add new actions primitives without major refactoring. I also have experience in working with APIs and app optimization that enables me to ensure a round-trip response time of less than 50 milliseconds. Lastly, I pride myself on delivering impeccable cross-platform applications and would be thrilled to leverage this expertise in Python and C++ bindings in Unity or Unreal to create an agent capable of autonomous thinking and behavior. My commitment to quality aligns perfectly with what you need; let's collaborate and build an impressive AI NPC together!
$140 AUD in 7 days
0.0
0.0

Hello, I understand that you require AI NPC automation and behavior development. You need an autonomous agent that reasons and acts through your SDK. I will build a decision-making layer using C++ and behavior trees for fast execution. With 5+ years of experience and a 5 star rating, I am confident that I can provide you with the best solution regarding NPC AI development. I assure you with professional communication until your work is done. Let's connect in chat so I can understand your requirements and get started right away. Best regards, Faisal Javed
$65 AUD in 1 day
0.0
0.0

As an experienced web and mobile developer, I’m no stranger to adapting to new technologies and platforms, which makes me a fantastic fit for your AI NPC project. Over my 9+ years of experience, I have undertaken many complex projects that involved real-time interactions. My knowledge in iOS, Android, Java, PHP,and HTML5 will be indispensable in handling the technical aspects of your SDK with swift round trip times. Moreover, being well-versed with Unity, I can integrate the existing observation/action interface into the decision-making layer seamlessly. Given my strong background in E-commerce and CMS based website development, I understand the importance of flexibility for future modifications and look forward to implementing reward function changes or tweaking difficulty without overhauling core logic. Additionally, I believe my keen eye for detail and structured approach make me an extraordinary asset for your project. Documentation and testability are paramount to successful software development - areas that I excel in. And not only do I offer competitive costs for my services, but as a member of your development team, I'll ensure you have extended support after project completion." If you're looking for an AI specialist who brings both dedication and adaptability to the table, let's bring your NPCs to life with fluid, intelligent actions using the power of artificial intelligence!
$140 AUD in 7 days
0.0
0.0

Hi, I can build the decision-making layer for your autonomous NPC agent and integrate it directly with the Python/C++ SDK observation and action interface. I’d recommend a hybrid approach using GOAP/behavior trees for reliable goal-driven behaviour, with adaptive/ML components where learning adds value, rather than hard-coding every action. I’ll structure it so the agent can process real-time state frames, choose context-aware actions, handle combat movement/weapon/defence decisions, social interactions and resource management, while keeping reward functions, difficulty and action primitives configurable. I’ll first validate everything in your headless harness, optimize the loop toward the <50 ms target, then test inside the 3-D environment. I’ll provide clean, documented, modular code with clear integration points and milestones.
$500 AUD in 7 days
0.0
0.0

Hi, Your POC needs a decision layer that can reason over live observations while remaining fast enough for real-time NPC control. I’d separate perception/state normalization, goal selection, decision-making and SDK action execution so behaviours can evolve from scripted policies toward learning-based agents without rewriting the integration layer. >>>Project Key Points: Python/C++ SDK integration, real-time observation parsing, combat tactics, strategic movement, weapon handling, defensive behavior, social interactions, resource management, adaptive/learning policies, <50 ms action loop, headless testing and Unity/Unreal-ready architecture. Execution Plan: I’d start with a modular hybrid architecture—GOAP/behavior logic for reliable goal-driven actions, with a learning component where adaptation adds measurable value. A normalized world-state model and action interface will allow reward functions, difficulty parameters and new primitives to be changed independently. I’ll benchmark the observation → decision → action pipeline early, then implement combat, social and resource-management behaviors incrementally. The headless harness will be used for deterministic regression tests before validating the agent inside the 3-D environment. Best regards, Prateek
$140 AUD in 7 days
0.0
0.0

Rockhampton, Australia
Payment method verified
Member since Aug 13, 2026
₹1500-12500 INR
$10-30 AUD
$250-750 USD
$30-250 USD
$100-400 NZD
₹12500-37500 INR
₹12500-37500 INR
€8-30 EUR
$5000-10000 USD
₹1500-12500 INR
$30-250 USD
$30-250 USD
€1000-3000 EUR
₹12500-37500 INR
₹750-1250 INR / hour
₹600-1500 INR
$10-30 USD
$25-50 USD / hour
₹750-1250 INR / hour
$80-200 HKD