Applied AI & Machine Learning Engineer

I build reliable AI systems for business workflows, patient access, sales intelligence, document understanding, and speech analysis.

My work focuses on turning complex or unstructured information—such as customer interactions, sales calls, contracts, catalogues, and administrative intake—into structured, reviewable, and actionable outputs.

Rather than treating the model call as the whole solution, I work on the engineering around it: ingestion, preprocessing, structured outputs, validation, uncertainty handling, persistence, APIs, testing, privacy boundaries, and human review.

I design applied AI systems that help teams reduce repetitive manual work, identify useful signals in complex data, and make AI outputs easier to inspect and use in real workflows.

My current work focuses on:

• AI agents and business automation
• Lead qualification and sales intelligence
• Patient access and administrative healthcare workflows
• Speech and sales-call analysis
• Document and PDF intelligence
• Vision-language model pipelines
• Retrieval-augmented generation
• Structured extraction from unstructured data
• AI reliability and uncertainty
• Backend and data-pipeline integration

A collection of applied AI systems built around business workflows.

Sales Call Analysis Agent
Transforms recorded sales conversations into structured, speaker-aware, and evidence-grounded performance insights through audio processing, transcription, diarization, speaker-role assignment, and analysis.

Catalog Vision Extractor
Processes visually complex PDF catalogues using vision-capable models and converts product and pricing information into validated, normalized Excel data.

Contract Analysis Agent
Analyzes contracts clause by clause and produces structured findings for obligations, risks, and supporting evidence through a local-first AI workflow.

A privacy-aware AI pipeline for turning authorized customer interactions into structured lead intelligence, qualification signals, and customer-care routing.

The project includes:

• authorized interaction ingestion
• lead and customer-care classification
• uncertainty escalation
• interest extraction
• catalogue-grounded retrieval
• tenant-aware persistence
• privacy and erasure foundations
• evidence and model-version tracking

The goal is to help businesses distinguish meaningful sales opportunities from support, irrelevant, and uncertain interactions while preserving clear privacy and access boundaries.

I am developing a set of connected, workflow-focused projects for patient access and clinic operations. They explore how administrative intake, staff interactions, follow-up, and operational review can fit together while keeping human oversight and data boundaries explicit. These repositories are private and the projects are in development; descriptions below reflect their current scope.

• Patient Journey Monitor — tracks patient communications and journeys, with SLA alerts and AI-assisted follow-up.
• AI Patient Intake & Referral Assistant — a synthetic-data pilot for answering approved sample FAQs, collecting administrative intake information, and creating a shared staff handoff. Staff can review transcripts and handoffs. It does not make LLM calls; hosted access is identity-based, and the pilot is not for real patient information.
• Patient Access Intelligence — a fictional-data, read-only analytics MVP that connects intake records, staff interactions, follow-up activity, and documented outcomes to surface evidence-backed operational signals and human-review priorities. It uses tenant-scoped synthetic records and deterministic classifications; it does not make clinical assessments, send messages, book appointments, or modify source systems. Real-data deployment remains blocked pending deployment-specific evidence and explicit approval.
• Integrated Access System — an integrated patient-access platform under development, intended to bring intake, booking, journey monitoring, staff support, and leakage analysis together. The current release focuses on the technical foundation: a React/TypeScript frontend, FastAPI backend, PostgreSQL and migrations, a background worker, and health/readiness checks. Patient, staff, and manager workflows are planned; the current interface checks the backend connection.

These are administrative and operational workflow projects, not clinical decision-making systems. Their current testing scope is fictional or synthetic data.

An ongoing research and technical-writing series exploring a broader question:

How can a model become more aware of when its own predictions may be unreliable?

The work covers:

• adversarial attacks and defenses
• uncertainty estimation
• out-of-distribution detection
• decision-boundary awareness
• architectural uncertainty signals
• evidential learning
• model robustness and selective prediction

The notebooks support the Humble Model Series, published through Medium and Towards AI. Recent repository updates include Essay #10 work, revised notebook material, and preparation for subsequent notebooks—beyond the Essay #8 companion notebook previously featured here.

An end-to-end regression application covering data preparation, model comparison, evaluation, saved-model inference, and Streamlit deployment.

Comparative classification experiments covering preprocessing, classical ML models, boosting methods, evaluation, and reusable model pipelines.

The projects below show earlier and supporting work across application prototypes, machine learning, data analysis, and visualization:

AI-powered application prototypes combining model integration, APIs, databases, and deployable interfaces.

Practical visualization studies using Iris and world-population datasets to explain relationships and distributions.

A regression study focused on hyperparameter tuning and overfitting.

Exploratory data analysis projects that use datasets to identify and communicate patterns.

A financial dashboard and analysis project. This repository is archived.

The ML Training and California House Price Predictor projects are described above. Together, these repositories show my progression from data analysis and classical machine learning toward applied AI systems and operational workflows.

I am particularly interested in problems where useful information exists but is difficult, repetitive, or expensive to process manually.

Examples include:

• turning complex PDFs and documents into structured data;
• analyzing large volumes of recorded sales conversations;
• identifying lead intent and prioritizing customer interactions;
• organizing administrative patient intake and follow-up workflows;
• separating sales opportunities from customer-care cases;
• grounding AI decisions in business catalogues or knowledge bases;
• designing RAG and rubric-based evaluation systems;
• integrating AI models with APIs, databases, and existing workflows;
• validating and structuring probabilistic model outputs;
• handling uncertain cases instead of forcing unreliable predictions;
• designing human-review and evidence-traceability mechanisms.

I try to build beyond the AI demo.

A typical workflow may involve:

Input → Validation → Preprocessing → AI Model → Structured Output → Verification → Persistence → Business Action

Depending on the system, I also design for:

• provider-independent model interfaces
• deterministic preprocessing
• schema validation
• retry and failure isolation
• confidence and uncertainty handling
• source evidence
• privacy-aware data handling
• automated testing
• static typing
• human review
• maintainable system boundaries

My goal is not simply to generate an AI response, but to make that response usable, inspectable, and reliable enough to become part of a real workflow.

Languages & Backend
Python · SQL · FastAPI · Pydantic · TypeScript · React

AI & Machine Learning
PyTorch · Transformers · Scikit-learn · NLP · RAG · AI Agents · Vision LLMs · Speech AI

Data & Persistence
PostgreSQL · SQLite · pgvector · SQLAlchemy · Alembic · pandas · Excel · JSON

Speech & Document Processing
faster-whisper · FFmpeg · FFprobe · PyMuPDF · PDF processing

Engineering
pytest · Ruff · mypy · Docker · Git · GitHub Actions · API integration

I write the Humble Model Series, exploring adversarial robustness, uncertainty, model awareness, and the limits of confidence in machine-learning systems.

📝 Medium — @maedehtorkian

🔬 Humble Model Series — Research Repository

• Email: maede.torkian@gmail.com
• LinkedIn: Maedeh Torkian
• Medium: @maedehtorkian
• Telegram: @maedehtorkian