Skip to content
BackNov 7
Gleb C.

Gleb C.

Product Engineer | Industrial Automation, AI Integration, Python, SQL

Product engineer: the product is my responsibility end to end, from turning a brief into a precise specification, through the build and the production launch, to the support that follows. I take on systems where the client needs a working result, not a developer to manage. AI & LLM systems engineering built on one architectural principle: deterministic core, LLM at the periphery. The model translates human intent and produces drafts--structured operations, SQL, estimates, documents--while validation, execution, and audit stay in deterministic code. Grounding is strict: every answer derives from the client's own data, every number is traceable to a source record with full provenance, never model-invented. Idempotency, at-least-once delivery with deduplication, event sourcing, eventual consistency across systems--the distributed-systems fundamentals--are engineered in from the start, not patched on later. Any case the system cannot ground is escalated to a human with a stated reason instead of a guess. What this covers in practice: RAG and knowledge-grounding pipelines built in-house without off-the-shelf frameworks (structure-aware ingestion, provenance tracking, content-hash idempotency--re-runs on unchanged inputs are zero-spend no-ops); hybrid lexical + embedding retrieval on PostgreSQL/pgvector; text-to-SQL and voice-enabled database agents with SELECT-only allowlists and query caching for deterministic repeat answers; AI drafting engines grounded strictly in the client's own historical data; leakage-audited evaluation with measured benchmarks (precision@1, task-phrased hit-rate); Anthropic Claude API, OpenAI API, batch-API routing for cost control. Selected AI projects are in my portfolio, including a public knowledge-base instance on GitHub. Alongside the AI work I remain, by training and by trade, a hands-on industrial automation and integration engineer with 15+ years in the field: PLC programming on Schneider and Siemens platforms, SCADA, OPC and Modbus connectivity, and SQL and Python integration between the plant floor and business systems. This work is still a large part of what I take on. Between the plant and the cloud I design and deliver production-grade middleware that synchronizes data between platforms using APIs and webhooks. Typical use cases include invoice generation, two-way status sync, Stripe webhook processing, controlled refund workflows via the Stripe Refund API (with human approval), SAP and ERP bridges, and a SQL-backed "sync state" layer (ID mapping, event journal, idempotency, retries, audit trails). The requirement behind all of them is stable automation that does not break the existing payment or operational flows. I also deliver enterprise-grade Seagull BarTender Enterprise solutions: designing production-ready .btw labels, integrating with SQL/ERP via ODBC and stored procedures, and deploying automated print workflows at scale. Using Integration Builder (REST/webhooks, TCP, file/queue triggers), Librarian (versioning/approvals), Administration Console and Print Portal (centralized deployment/permissions), and History Explorer/System Database (audit trails, reprints, enterprise-wide archiving), I build robust, compliant labeling systems suitable for multi-site operations. For ERP/WMS-driven environments, I implement BarTender REST API integrations (JSON/YAML over HTTP or HTTPS, Windows/Basic auth, IIS+Print Portal setup), enabling external systems to trigger on-demand label printing directly without file-based intermediaries--complementing Integration Builder for scenarios where the calling system already speaks REST. Core stack: Schneider Modicon (M2xx/M340/M580) programmed in Unity Pro / EcoStruxure Control Expert and Machine Expert (IEC 61131-3: Function Block, Structured Text, Ladder), Siemens TIA Portal (v20) across SIMATIC S7-1200/S7-1500 and legacy S7-300/400 in STEP 7 (SCL, LAD, FBD, STL), ET 200SP/ET 200M distributed I/O, WinCC (Basic/Comfort/Advanced and WinCC Unified), Startdrive for SINAMICS drives, HMI (Schneider Magelis/Harmony with Vijeo Designer; Siemens KTP/TP Comfort series in TIA), SCADA, OPC UA/DA, PROFINET/PROFIBUS DP, Modbus RTU/TCP, CANopen, PostgreSQL/pgvector, Obsidian, Anthropic Claude API, OpenAI API, SQL Server, Python (FastAPI/Flask/pyodbc), Windows & Linux deployment (NSSM, services, nginx). Labeling & enterprise printing: Seagull BarTender Enterprise (Integration Builder, Librarian, Administration Console, Print Portal, History Explorer/System DB), SQL/ODBC, ERP/OPC/REST bridges, automated print and archival at scale. Printers: SATO, Zebra (ZPL II), TSC, Honeywell/Intermec--thermal transfer and direct thermal, print-and-apply applicators. Barcode and RFID encoding: GS1-128, GS1 DataMatrix, GS1 QR, GS1 Digital Link, SSCC and GTIN application identifiers, EPC/RAIN RFID tag encoding (SGTIN-96, GRAI) for UHF smart labels.

🇩🇪Berlin, Germany$65/hr100% JSS4.99 (10)Since Sep 2025

Earnings and projects

MRR
$1.8k

Last 6 months ÷ 6

World#25,504 / 412k
🇩🇪Country#191 / 3,842
Revenue, last 30 days
$1.6k

Last 6 months: $11k

World#28,510 / 412k
🇩🇪Country#198 / 3,842
Total earnings
$16k

Billed in all: $17k

World#157,212 / 412k
🇩🇪Country#1,209 / 3,842
Per project
$1.3k

$3.6k in the last 6 months

World#102,327 / 412k
🇩🇪Country#803 / 3,842
Projects, last 6 months
3
67% fixed33% hourly
World#66,479 / 412k
🇩🇪Country#535 / 3,842
Projects in all
12
67% fixed33% hourly
World#183,649 / 412k
🇩🇪Country#1,585 / 3,842

Earnings, last 30 days

$16.5k+$1.6k earned in the last 30 daysBilled in all, before Upwork's fee · drag across the chart to see what was earned in any stretch
$14k$15k$16k$17kOct 8Oct 23Nov 7

Skill rankings: #1 in 1 skill of 20

Rarest: top 5% in Data Management, of 2,504. Ranked by MRR among everyone on Upwork with the skill.

  1. 1Seagull Scientific BarTender6 freelancers · 🇩🇪 #1Top 17%
  2. #12SCADA128 freelancers · 🇩🇪 #1Top 9%
  3. #15PLC Programming290 freelancers · 🇩🇪 #1Top 5%
  4. #18Industrial Automation173 freelancers · 🇩🇪 #2Top 10%
  5. #50Big Data559 freelancers · 🇩🇪 #2Top 9%
  6. #118Data Modeling1,203 freelancers · 🇩🇪 #2Top 10%
  7. #129Data Management2,504 freelancers · 🇩🇪 #3Top 5%
  8. #194Data Engineering1,602 freelancers · 🇩🇪 #2Top 12%