AI SOLUTIONS ENGINEER | GENAI · RAG · AI AGENTS · MLOPS
I bridge machine learning and production engineering: turning modern AI capabilities—from RAG architectures and autonomous agents to computer vision—into resilient, scalable cloud systems that solve genuine operational challenges.
02 / ABOUT & METHODOLOGY
Operational engineering experience applied directly to the frontier of artificial intelligence.
I am an AI Solutions Engineer grounded in over 7 years of infrastructure, cloud architecture, automation, and DevOps experience.
AI solutions cannot live in an isolated notebook: they require robust data ingestion, fast vector retrieval, resilient APIs, secure authentication, containerized orchestration, and rigorous telemetry in production.
My progression connects the entire engineering stack: Python → Machine Learning → APIs → LLMs → RAG → Vector Databases → AI Agents → Evaluation → MLOps. Beyond building software, I teach applied AI at ReDI School of Digital Integration, serve as Adjunct Faculty at Cox's Bazar International University, and co-founded MoviSabio to advance intelligent transportation systems.
HOW I BUILD AI SOLUTIONS · 6-STAGE LIFECYCLE
INTERACTIVE AGENT · LOCAL CLIENT-SIDE RETRIEVAL
Explore my engineering work, technical expertise, projects, and professional background through a private, interactive knowledge interface.
Answers are computed 100% locally in your browser using deterministic layered retrieval: intent classification, entity & alias resolution, TF-IDF scoring, and structured composition grounded strictly in Alimul's 27 documented projects and career history. No remote AI or embedding API is called, and no visitor data leaves your device.
Hi! I am Alimul's AI Twin, running locally in your browser through a deterministic knowledge retrieval engine. Ask me anything about Alimul's 27 verified projects, Enterprise RAG architectures, Azure & Terraform infrastructure, computer vision systems, or engineering qualifications.
Click any of the suggested prompts below, ask for project comparisons, or paste a target job description for a grounded suitability analysis!
03 / FLAGSHIP INITIATIVES
Deep dive into flagship production reference platforms and live engineering initiatives.
A production-grade, multi-tenant Retrieval-Augmented Generation (RAG) reference architecture built to eliminate hallucinations and unlock institutional knowledge. Implements FastAPI, PostgreSQL with pgvector, Reciprocal Rank Fusion (RRF), cross-encoder reranking, hierarchical document chunking, and tenant-scoped RBAC controls.
A territorial intelligence initiative optimizing urban traffic flow and reducing intersection wait times. Integrates geospatial data processing, vehicle volume analytics via computer vision, and microscopic traffic simulation models (SUMO/TraCI, CARLA, PostGIS) for dynamic signal coordination.
ENGINEERING IMPACT · BEFORE VS AFTER
See how disciplined engineering solves baseline limitations across flagship RAG, perception, and clinical machine learning platforms.
INTERACTIVE TRADEOFF CHALLENGE
Test your engineering judgment on a real-world dilemma: balancing strict hallucination suppression with sub-second production latency.
Your Enterprise RAG platform is retrieving verified knowledge from 250,000 internal technical manuals. However, during ambiguous user queries, the LLM occasionally fills knowledge gaps with plausible hallucinations. The Product Director demands zero hallucinations, while the VP of Engineering mandates maintaining < 650ms p95 latency. How do you engineer the solution?
04 / PROJECT DIRECTORY
Filter by domain to inspect dedicated repositories, machine learning prototypes, computer vision systems, and cloud infrastructure code.
A computer-vision roadway perception pipeline with CARLA simulation integration. Performs RGB preprocessing, region-of-interest masking, Canny edge detection, Hough-line extraction, left/right lane-boundary estimation, and temporal smoothing for driverless simulation testing.
A leakage-free, supervised classification system trained on the UCI Cleveland dataset. Adheres to enterprise constraints using permitted model families (Logistic Regression, Decision Trees, Random Forest, Soft Voting Ensemble) paired with a containerized FastAPI service, database audit logging, and CI quality checks.
An enterprise-grade, modular Infrastructure as Code (IaC) deployment framework for Azure. Automates multi-tier cloud provisioning (Virtual Networks, Subnets, Key Vaults, Kubernetes clusters) using reusable Terraform modules and passwordless OIDC-authenticated GitHub Actions pipelines.
Search and filter all 27 verified technical systems across Embedded Robotics, Computer Vision, CARLA Sim, Data Engineering, Azure Cloud, and Generative AI.
Explore Full Project Archive (27 Projects)MEDIA SHOWCASE
Explore selected moments from engineering projects, robotics experiments, drone and travel photography, technical demonstrations, and visual explorations.
05 / CAREER JOURNEY
A track record spanning AI instruction, technical leadership, and 7+ years of infrastructure operations.
06 / ACADEMIC & CREDENTIALS
Formal engineering education, peer-reviewed international publications, and verified professional certifications.
Active examination preparation for Azure Administrator (AZ-104), Azure AI Engineer (AI-102), Azure Developer (AZ-204), and Azure Solutions Architect (AZ-305).
07 / TECHNICAL CAPABILITIES
Technical proficiencies organized by discipline: AI engineering, core software, cloud infrastructure, and intelligent systems.
Interdisciplinary proficiencies bridging automated workflows, business intelligence analytics, and industrial hardware-level engineering.
08 / PROFESSIONAL SERVICES CATALOG
Comprehensive production engineering across 9 technical domains and 95 specialized services. From bespoke web applications and enterprise management systems to custom Generative AI agents, RAG knowledge retrieval, and resilient cloud architectures.
TRANSPARENT PRICING / EFFORT MODELING
Calculate preliminary engineering effort, sprint delivery timelines, and budget ranges across USD, BDT, and EUR. Transparent pricing formulas with separate recurring maintenance and third-party operational cost projections.
AUTOMATED SCOPE SYNTHESIS / FORMAL DOCUMENT GENERATION
Transform operational objectives and technical criteria into a formal, 21-part engineering proposal complete with architectural breakdown, milestone schedule, pricing milestones, and SLA terms.
Mode B Hook Active: In-browser AI drafting with client-side API execution. Your API key is stored only in memory for this session and is never logged, transmitted to any backend, or persisted.
09 / INITIATE CONTACT
Whether you are looking to architect enterprise RAG platforms, deploy autonomous AI agents, scale cloud-native infrastructure, or collaborate on technical challenges—my inbox is open.