Case studies
13 best developments
Models kept and cut, the papers behind them, the architecture, and what ran in production. Demo recordings where they exist.
-
Askimo
Askimo — EvoTalk Voice Agent
Multilingual voice agent for SME sales operations, with live CRM context injected mid-call and sub-second end-to-end latency held across roughly twenty concurrent conversations.
- Python (FastAPI, asyncio)
- Sarvam Models
- Mistral 7B Instruct
- Silero VAD
-
925 Worker LTD
RAG System for Clinics and Patient Data Analysis, CRM and Scheduling Integration
Retrieval-augmented clinical question answering: every response is generated from retrieved literature, cites its sources, and is fact-checked against that same evidence before it reaches a clinician.
- Python (FastAPI, asyncio)
- Llama 3.1 8B Instruct
- GPT-4 Turbo (fact-check path)
- Sentence Transformers (all-MiniLM-L6-v2)
-
DigitB (Freelance)
Multi-Agentic Memecoin Generation System
A team of specialised agents that argues its way to a token concept, commissions the artwork, writes the launch post and deploys the token on-chain — one operator prompt in, a live token and a published thread out.
- Python (asyncio)
- Multi-agent orchestration
- LLM tool / function calling
- DALL·E image generation
Demo & stills -
Mountaingreen
Mountaingreen.at — Platform Migration & Data Recovery
Rebuilt an affiliate platform's backend off deprecated JavaScript frameworks onto a modern Node.js/Express stack with a proper routing layer, migrated it from a single hand-configured EC2 box onto ECS Fargate behind Terraform, and reconstructed 2,347 orders that a failed database migration had silently dropped two years earlier.
- Node.js (Express, modular OOP)
- React 18 (TypeScript, Vite)
- PostgreSQL 16 (RDS, read replicas)
- Redis (ElastiCache)
-
DigitB (Freelance)
Solana Multi-Agent DeFi Framework
A multi-agent framework for the Solana ecosystem — specialised agent workflows for token generation, contract deployment and exchange integration, driven from a browser console that streams every agent's output as it works.
- Python (asyncio)
- Multi-agent orchestration
- LLM tool / function calling
- Solana (token generation, contract deployment)
Demo & stills -
Chimera Compound LLC
Chimera Studio — Product Image Synthesis
Text-driven product photography editing and generation for e-commerce — the platform launched as LayFlat AI and later renamed Chimera Studio — turning a single garment photo into studio-grade flat-lay imagery.
- PyTorch / Hugging Face Diffusers
- SegFormer
- ControlNet
- InstructPix2Pix
Demo & stills -
DigitB (Freelance)
StyleTTS 2 Fine-Tune — Long-Form Narration Voice
A single-speaker narration voice built by full fine-tuning of StyleTTS 2, for a daily Torah-study publication whose English text is dense with transliterated Hebrew and Aramaic. Corpus construction under an SNR gate, a staged 100-epoch training run on rented GPU, and acceptance on an 8 min 41 s continuous generation rather than on cherry-picked clips.
- Python / PyTorch
- StyleTTS 2
- PL-BERT
- HiFi-GAN
Demo & stills -
Zenfinity
Zenfinity — Conversational Astrology Platform
StarSage is the Vedic-astrology AI at Zenfinity, a pre-seed AI wellness platform. In-depth astrological conversations from agents grounded in nakshatra, dasha and live planetary data, served from fine-tuned open-weight models rather than a frontier API.
- Python (FastAPI, asyncio)
- Llama 3.1 8B Instruct
- Mistral 7B Instruct
- Hugging Face Transformers / PEFT
Demo & stills -
DigitB (Freelance)
Real-Time ATC Communication System
Live transcription of air-traffic-control radio channels: UDP audio transport straight into Whisper, with the resulting text rendered against a sector map as it arrives.
- Python
- OpenAI Whisper (speech-to-text)
- UDP socket transport
- Real-time audio conditioning and segmentation
Demo & stills -
DigitB (Freelance)
Intelligent Educational Voice Assistant
A spoken-language tutor that answers a student's question out loud by retrieving the relevant passage from NCERT school textbooks first, then explaining it — retrieval-augmented generation wrapped in a voice loop.
- Python (FastAPI, asyncio)
- Whisper (speech-to-text)
- Sentence Transformers (all-MiniLM-L6-v2)
- Mistral 7B / Llama 3.1 8B
Demo & stills -
VocalEats
VocalEats — Restaurant Voice Ordering Agent
Voice agents that take restaurant orders over a real phone line and write them straight into the Square POS — WebRTC transport, ASR, intent and entity extraction, payment capture and human handoff.
- Python (FastAPI, asyncio)
- Whisper (speech-to-text)
- Mistral 7B (intent + entity extraction)
- LiveKit (WebRTC)
-
MarketingHub.ca (Dave Groups Inc, Ontario)
MarketingHub.ca — LLM Automation for a Digital Agency
A long-running engagement building LLM-backed automations into an agency's existing stack — HubSpot, Zapier, Monday.com, Slack — with a human approval step on every output that reaches a client.
- Python (FastAPI, asyncio)
- Mistral 7B Instruct
- Llama 3.1 8B Instruct
- GPT-4 (selective)
-
DigitB (Freelance)
Cryptocurrency Analysis & Trading Framework
Modular Python framework for real-time cryptocurrency analysis: multi-exchange price feeds, vectorised technical indicators, sentiment aggregation and a backtesting engine that validates a strategy before any capital is at risk.
- Python (FastAPI, asyncio, multiprocessing)
- NumPy / Pandas (vectorised computation)
- TextBlob / VADER (sentiment)
- WebSocket (real-time exchange feed)