Freelance AI Agent Orchestrators

Freelance AI Agent OrchestratorsFreelance AI Agent Orchestrators

Projects People

Keaton Williams

San Francisco, USA

AI Automation Consultant & Strategist

1x: Hired

18: Followers

Expert

AI Automation Consultant & Strategist

Cover image for Numun.io AI Platform Overview

Numun.io AI Platform Overview

1

12

Cover image for In Numun.io (http://Numun.io), workflows are first designed ...

In Numun.io (http://Numun.io), workflows are first designed and validated in Agent Studio as structured, multi-step graphs—made up of agents, tools, conditionals, loops, and outputs. Once a workflow is complete, it doesn’t live in isolation. By toggling the workflow on in Agent+ Mode, users can execute that same workflow directly from the AI chat interface. Instead of receiving a simple text response, the AI runs the full workflow behind the scenes and returns structured run results, including tool outputs, decisions taken, and final outcomes. The result is an AI experience that goes beyond conversation, turning chat into a live interface for executing and inspecting real autonomous workflows. In this example, Python code is generated and executed via Fly.io (http://Fly.io), returning a structured summary of results.

20

430

Cover image for AI Builder + Agent Studio is a system for designing, testing...

AI Builder + Agent Studio is a system for designing, testing, and deploying AI agents and autonomous workflows as structured, production-ready graphs, not fragile prompts. At the core is Agent Studio, where workflows are defined as explicit graphs made up of agents, tools, conditionals, loops, routers, and outputs. The AI Builder sits on top of this system and generates complete workflows automatically. Instead of writing code or wiring nodes manually, users describe the goal—and the AI Builder constructs a valid workflow graph. Every workflow can be tested using a test harness, ensuring agents execute correctly before deployment. This project demonstrates how AI agents can be treated like real software systems—versioned, validated, and orchestrated—instead of one-off prompt chains.

26

509

Cover image for Numun.io (http://Numun.io) is an autonomous agent and workfl...

Numun.io (http://Numun.io) is an autonomous agent and workflow execution platform designed to turn intent into action—reliably, securely, and at scale. Numun enables users and teams to design, deploy, and run AI-powered agents and workflows that browse the web, call tools, orchestrate systems, and complete real tasks end to end. At its core is a contract-based execution model that emphasizes determinism, observability, and accountability—bringing production-grade discipline to AI agents. Numun is built for developers, operators, and founders who need agents that actually work in the real world—not demos that break under pressure. This project includes full-stack platform development, execution infrastructure, agent runtime design, workflow orchestration, and system-level reliability engineering.

23

429

AI Agent Orchestrator(3)

Armughan Shahid

proIslamabad, Pakistan

AI SaaS Dev | LLMs, Agents, Voice & Automation | Web, Mobile

New to Contra

AI SaaS Dev | LLMs, Agents, Voice & Automation | Web, Mobile

Cover image for AI Operations Agent: RAG-Powered Retail

AI Operations Agent: RAG-Powered Retail Intelligence & Task Automation This project was built for large-scale restaurant groups and multi-unit retail operators who manage high volumes of data across dozens or hundreds of locations. Specifically designed for Regional Managers and Operations Directors, the system serves as an enterprise-grade "Digital Consultant" that bridges the gap between fragmented POS/inventory data and daily on-the-ground execution. By transforming millions of rows of restaurant performance metrics into high-priority tasks, it provides a centralized platform for leadership to monitor KPIs, approve AI-suggested corrective actions, and ensure operational consistency across their entire portfolio. 1. What We Built We developed a production-ready Autonomous AI Operations Agent designed to bridge the gap between complex retail data analysis and daily execution. The system acts as a digital consultant for regional managers, transforming raw KPIs into actionable tasks. Analytical AI Chat: A free-form conversational interface where users can query performance data (e.g., "Show me the top 5 worst profitable stores in Istanbul for the last 3 months"). Task Management Dashboard: A structured workflow where AI-suggested actions are automatically logged for manager approval or rejection. Automated Action Logic: The agent uses an "Action Suggestion Map" to identify specific defects (like low audit scores or high food waste) and suggest precise corrective measures. Persistent Memory: Includes both short-term memory for the current chat session and long-term RAG memory to maintain context over time. 2. How We Built It (The Stack) The system was engineered for scalability and reliability using a modern, containerized stack:AI Orchestration: LangGraph was used to manage complex, multi-turn reasoning and agentic workflows. Frontend: React/Next.js 14 for a responsive, real-time user interface. Backend & Data: Node.js paired with a PostgreSQL database capable of handling 1M+ records. LLM Access: Integrated via OpenRouter to allow for flexible model selection and switching. Infrastructure: Fully Dockerized to ensure consistent deployment across environments. 3. Challenges We Faced As the system scaled from prototype to processing millions of records, we encountered several critical engineering hurdles: Response Latency: The initial monolithic prompt architecture led to response times exceeding 60 seconds, far slower than the required "ChatGPT-like" speed. Prompt Verbosity & Errors: Complex questions involving multiple variables caused the LLM to lose focus, leading to "reasoning errors" and incorrect SQL generation. Hallucination Risks: In multi-branch queries, the model occasionally fabricated data points, particularly around manager hours and performance metrics. Context Switching Bugs: The agent sometimes struggled to "let go" of a previous topic, continuing to reference an old store when the user had asked about a new city. 4. How We Solved It We re-engineered the core pipeline to transition from a single, heavy agent into a Modular Multi-Step Architecture: 75% Latency Reduction: By decomposing the main logic into smaller, task-specific nodes, we dropped processing time from 60s down to 15s. Task Decomposition & Specialized Models: We stopped using a "one-size-fits-all" model. Instead, we implemented a router that uses lighter, specialized models for SQL generation and action identification, and flagship models only for final reasoning. Granular SQL Generation: Breaking the metadata analysis into narrow sub-steps eliminated SQL hallucinations. The model now only "sees" the specific schema needed for the current sub-task, ensuring 100% accuracy. 10-Point Testing Protocol: We implemented a rigorous QA protocol that specifically verified bug fixes for context switching, task duplication, and chart coverage before final delivery.

0

106

Cover image for BudgetNest — AI-Powered Personal Finance

BudgetNest — AI-Powered Personal Finance SaaS Most people don't track their finances because the friction is too high. BudgetNest removes that friction entirely, every transaction captured automatically, categorised intelligently, and surfaced through analytics that actually help people make better decisions. The core problem it solves: Manual expense logging fails because people forget, get lazy, or simply don't have time. BudgetNest built an automated capture layer that works across every channel a user already operates in i.e. SMS alerts, bank emails, receipt photos, WhatsApp messages, and voice notes in English and Urdu. The system deduplicates intelligently across all input sources so nothing gets logged twice regardless of how it came in. What was built: A complete AI finance platform with five distinct automated capture modes SMS and email parsing for bank transaction alerts, PDF and image bank statement upload with AI extraction, OCR receipt scanning via camera, a WhatsApp bot that accepts text, images, and voice notes, and multilingual voice input for manual cash payments. Every transaction flows through an LLM-powered categorisation engine that auto-assigns categories and subcategories, recognises vendors, and learns from behaviour over time. Beyond capture, the system includes smart budgeting with AI-driven suggestions based on spending patterns, subscription detection for recurring transactions, shared expense and split-bill tracking, fraud detection for unusual transactions, and forecasting that projects deficit against income. Dashboards surface everything through charts, trend lines, and weekly and monthly summaries. Technical architecture: React Native across iOS and Android, Node.js and FastAPI backend, PostgreSQL and MongoDB, AWS infrastructure with EC2, S3, and RDS, Python-based NLP and OCR pipeline using Transformers and Tesseract, Twilio WhatsApp integration, Gmail API for email parsing, and Firebase for push notifications. Business model built in from day one: Freemium with premium automation features, B2B white-label capability for microfinance institutions and NGOs, and the OCR and SMS parsing logic architected as standalone APIs for third-party licensing meaning the AI layer has revenue potential independent of the consumer app.

1

129

Cover image for We Step Together - Step-to-Donation Mobile App

We Step Together - Step-to-Donation Mobile App

0

1

Cover image for Development of CarlsbergHub & ParkingHub

Development of CarlsbergHub & ParkingHub

0

1

AI Agent Orchestrator(2)

proMálaga, Spain

Full Stack Developer | AI Product Developer

$1k+: Earned

2x: Hired

5.0: Rating

30: Followers

Expert

Expert

+1

Full Stack Developer | AI Product Developer

Cover image for Agentic AI Persona Chat (Replit)
Agentic

Agentic AI Persona Chat (Replit) Agentic AI Persona Chat (Replit) is an interactive AI chat application built on Replit that enables users to converse with specialized AI personas—such as a Helpful Assistant or Socratic guide—designed to reason, guide, and act autonomously based on user intent. The project showcases agentic behavior, modular personas, and real-time conversational intelligence in a developer-friendly environment.

2

178

Cover image for Day 23/30 Everyone’s still chasing 1000s of trash leads like...

Day 23/30 Everyone’s still chasing 1000s of trash leads like it’s 2020. Wrong. Lead gen died with volume. It lives with precision. I built a zero-human daily machine in n8n that outperforms a junior SDR + researcher + VA — and costs pennies: PhantomBuster → scrapes fresh LinkedIn leads on exact target keywords → auto cleans & dedupes → AI qualifies (role + relevance + buying intent) → drops only winners straight into Notion → fires hyper-personalized cold emails instantly → pings me in Slack so ZERO fall through cracks No Google Sheets hell. No copy-paste torture. No spraying & praying.Just a silent, ruthless lead engine that runs every morning while I drink coffee. If you’re still manually hunting leads in 2026… you’re not early anymore. You’re late. Drop 🔥 if this hits.#30Days30Automations #BuildInPublic #LeadGeneration

3

245

Cover image for Agentic AI Persona Chat (Replit)
Agentic

Agentic AI Persona Chat (Replit) Agentic AI Persona Chat (Replit) is an interactive AI chat application built on Replit that enables users to converse with specialized AI personas—such as a Helpful Assistant or Socratic guide—designed to reason, guide, and act autonomously based on user intent. The project showcases agentic behavior, modular personas, and real-time conversational intelligence in a developer-friendly environment.

2

190

Cover image for Workflow Automation and RAG Agent System for Canal Digital

Workflow Automation and RAG Agent System for Canal Digital

2

14

AI Agent Orchestrator(2)

proDublin, Ireland

Ping me from the edge of innovation | CTO at RaptorLabs.dev

$1k+: Earned

1x: Hired

5.0: Rating

296: Followers

Expert

Ping me from the edge of innovation | CTO at RaptorLabs.dev

Cover image for This is not your typical

This is not your typical hackathon. Notion + AI agents opens up a completely new way of building workflows. If you’re serious about building, this is the one. Join here: https://on.contra.com/q6DNjf 🚀 @Notion x @Contra #notioncustomagentbuildathon (https://www.linkedin.com/search/results/all/?keywords=%23notioncustomagentbuildathon&origin=HASH_TAG_FROM_FEED)

1

10

553

Cover image for PROJECT NAME
AION PYR — The

PROJECT NAME AION PYR — The Aitherioi PROJECT LINK https://app.melius.com/projects/029a5c7f-3a2a-4c7b-a24c-e25f36815d91/canvas/98b87dac-fcc4-411b-8695-83b7dc318726 CONCEPT / PROJECT DESCRIPTION AION PYR is a one-minute instrumental progressive-rock film by The Aitherioi, a fictional trio of pale pre-terrestrial beings older than Earth. They were here when the planet was born from heat, basalt, and lava, and they remain calm as that world dissolves into stardust. The film follows them performing inside a collapsing volcanic cathedral while lava, steam, lightning, and stone slowly give way to void. The music is built only from guitar, bass, and drums: slow, heavy, repetitive, bass-led, and cathartic. PROCESS I started by defining The Aitherioi: a fictional pre-terrestrial trio with a shared visual identity, pale ethereal bodies, severe faces, long dark hair, and calm expressions. Then I built the visual system around contrast: fossil-dark basalt, living lava, white steam, lightning, and their stillness inside collapse. I created character assets, face studies, wardrobe references, instrument references, environment plates, keyframes, and individual video scenes. The music was built first as a strict one-minute instrumental progressive-rock track using only guitar, bass, and drums. The video was then structured to follow the track: bass opening, wide trio, basalt corridor, guitar catharsis, drummer pulse, matter dissolution, and final void. Finally, I stitched the scenes together with clean cuts so the film follows the music without extra transitions or title cards. FEEDBACK ON BUILDING WITH MELIUS Melius worked best as a production canvas rather than a single-prompt generator. The node-based workflow made it possible to build the project in layers: character identity, environments, audio, keyframes, video clips, and final assembly. That helped keep the concept coherent while still allowing corrections when individual shots needed refinement. The biggest challenge was visual continuity, especially keeping The Aitherioi consistent across scenes. The most useful approach was creating strict reference assets first, then using them as anchors for keyframes and video generation.

18

31

1.7K

Cover image for Just published a new case

Just published a new case study on Contra: IntelliChance, an AI-powered Solana explorer website built with v0.app (https://v0.app/ref/UMDY33) and deployed on Vercel. Dark editorial UI, amber-black visual identity, responsive design, and a premium Web3 product narrative. Case study: Contra Link (https://contra.com/p/Ok3gqcUR-intelli-chance-ai-powered-solana-explorer-website)

16

20

802

Cover image for IntelliChance: AI-Powered Solana Explorer Website

IntelliChance: AI-Powered Solana Explorer Website

11

8

AI Agent Orchestrator(6)

proZagreb, Croatia

AI-Native Builder | Apps, Templates & Websites Shipped Fast

1x: Hired

5.0: Rating

183: Followers

Expert

AI-Native Builder | Apps, Templates & Websites Shipped Fast

Cover image for This is not a portfolio.

This is not a portfolio. ⚡ It is a live personal system built during the challenge window. Yels is my personal website and control layer, built and hosted entirely on Zo Computer. It combines: 🟣 real-time Solana data 🔐 external Telegram control — /lock /unlock 💳 direct Stripe conversion 📊 hidden layers beneath articles and signal charts Creative vision: to turn a personal website into something alive — part identity layer, part data terminal, part monetization system, and part hidden interactive experience. Everything is live. 🟢 Built on my Zo. Live site: https://yels.zocomputer.pro (https://yels.zocomputer.pro)Video walkthrough: https://youtu.be/Hjt3HmK79cs

8

17

782

Cover image for 🏆 The Zo Computer Challenge

🏆 The Zo Computer Challenge is an amazing opportunity and the perfect time to build something genuinely original. Get hands-on with OpenClaw without dealing with a complicated setup. Don't miss this one! 🔥 https://on.contra.com/QpQmPg 👈 @Zo Computer & @Contra supercombo! ♥️ 💰 $10,500 in prizes for creating the wildest personal website on the internet. #zocomputerchallenge @Ben Guo and rest of the Zo team did amazing job!

10

289

Cover image for Building on Zo isn't just

Building on Zo isn't just work - it's pure fun. The flow is so addictive that I’m officially down the rabbit hole and I'm not coming out. Build on Zo here: 👉 https://on.contra.com/QpQmPg

4

17

1.1K

Cover image for The machine had seen invention,

The machine had seen invention, war, fame, and gravity. Music was the first thing it did not understand. @CapCut x @Contra HQ

5

14

1.6K

AI Agent Orchestrator(3)

proSan Francisco, USA

AI & ML Engineer

$1k+: Earned

7x: Hired

24: Followers

AI & ML Engineer

Cover image for Claude Chat Assistant
A modern real-time

Claude Chat Assistant A modern real-time chat messaging application with integrated Claude AI assistance. Built with Next.js 15, React 19, Supabase, and Tailwind CSS. Features: Claude AI Integration - Chat with Claude AI for instant assistance with writing, coding, brainstorming, and more Private & Group Chats - Create one-on-one conversations or group chats for team collaboration Real-time Messaging - Messages appear instantly with Supabase real-time subscriptions Rich Media Support - Send images, files, and emojis in your conversations Read Receipts - See when your messages have been read Message Actions - Reply to, edit, or delete messages User Presence - See who's online, away, or offline Dark Mode - Toggle between light and dark themes Responsive Design - Works beautifully on desktop and mobile devices

0

120

Cover image for AI Groceries
An AI-powered online grocery

AI Groceries An AI-powered online grocery delivery platform built with Next.js 15, React 19, Supabase, Claude API, and Stripe. Designed around "The Harvest Table" philosophy -- abundant, warm, and grounded in real food provenance.

1

183

Cover image for Biglabs
Consulting-grade template for professional services

Biglabs Consulting-grade template for professional services firms. Manage case studies, articles, services, careers, testimonials, and client consultations with AI-powered analytics. Deploy in minutes, no local setup required. Features Thought Leadership — Publish articles and insights with categories, excerpts, and rich content Service Pages — Present your consulting services with detailed descriptions and feature lists Consultation Booking — Public intake form for prospective clients with scheduling preferences Careers Portal — Post open positions with departments, requirements, benefits, and salary ranges AI Analytics — Claude-powered insights and reporting on your consulting data And more.....

0

188

Cover image for AI Vested
An AI-powered investment platform

AI Vested An AI-powered investment platform with Claude AI integration. Deploy in minutes, no local setup required.

0

213

AI Agent Orchestrator(1)

proMumbai, India

Product studio building with early-stage founders.

New to Contra

Product studio building with early-stage founders.

Cover image for Mr. Loop: Intelligent Onboarding Agent

Mr. Loop: Intelligent Onboarding Agent

0

1

Cover image for Hey!
Here is a looping

Hey! Here is a looping motion graphics video that we created for the hero block of a website. With 3 key words Vintage + Nostalgia + Energy, this was the outcome!

0

113

Fitness Brand Landing Page Design

0

1

Cover image for Innovative Global News Platform: Streamlined Design & UX Ins...

Sharing a recent project - Newsroom Newsroom is a global news and publication platform I designed end-to-end. The idea was simple: create a single destination for discovering news across sources, without the overwhelm. Newsroom aggregates content from a wide range of publications, covering current affairs, sports, entertainment, and more, designed for a global, diverse audience. 🧠 Key UX decisions 1. Swipe-based interaction for discovery I introduced a simple swipe gesture on news cards: → Swipe right to like → Swipe left to skip This made browsing feel lightweight and intuitive, especially when dealing with high volumes of content. It also subtly personalizes the feed over time. 2. Reducing information overload Instead of dense lists, I leaned into a card-based layouts with clear hierarchy, allowing users to scan quickly and decide what to engage with. 3. Designed for global consumption I focused on a dark theme, with clean visual language and flexible layouts that work across categories and regions, without bias toward any one type of content. ⚙️ What I worked on •⁠ ⁠End-to-end product design •⁠ ⁠Core discovery and reading flows •⁠ ⁠Design system for scalability across categories •⁠ ⁠Interaction patterns for engagement and personalization The goal was to make news consumption feel less overwhelming, more intentional, and easier to navigate. Would love to hear your thoughts!

1

283

AI Agent Orchestrator(1)

United Arab Emirates

Award-winner impact focused UI UX Designer | Product Design

5.0: Rating

6: Followers

Award-winner impact focused UI UX Designer | Product Design

Cover image for Innovative AI-Driven Interface for 2030 Sustainable Goals

The 17 Sustainable Development Goals are essentially the world’s to-do list for 2030. I wanted to build an immersive interface that actually lets countries and governments interact with their progress. It’s built for deep dives where you can compare performance across borders, analyze raw data, and even use a policy sandbox to "dummy try" initiatives to see simulated results before they happen. 🚀 Honestly didn’t think I could cram all of this in, but Figmake helped me pull off features and intuitive experience. I managed to build an AI-driven story mode, conversational inputs, and :a unique custom circular filter. Started same day, around 200 regenerations, last minute submission; please accept! #FigmaMake #SDG #GlobalGoals #UIUX #DesignChallenge Figma Site - https://many-scarf-95159460.figma.site

4

18

844

Cover image for Product Discovery Workshops for SaaS, CRMs, ERP and MVPs

Product Discovery Workshops for SaaS, CRMs, ERP and MVPs

0

16

Cover image for Synavos Rebrand

Synavos Rebrand

0

5

Nitro League - Brand Identity, Website & Game Experience Design

0

9

AI Agent Orchestrator(1)