Rajmohan Bajaj

As the client-facing AI Engineer at GNX Group I transformed high-level product ideas into fully deployed, revenue-generating solutions built on multi-agent frameworks. Partnering directly with enterprise stakeholders, I scoped technical requirements, designed agentic architectures, and led iterative delivery sprints—from prompt-engineering and model evaluation. A flagship outcome was an LLM-driven QA-Automation suite for static websites and complex web-apps. The pipeline orchestrates specialised agents that: (1) crawl a site’s DOM to map user journeys, (2) auto-draft human-readable test cases, (3) convert them into Playwright test scripts, and (4) execute and self-verify results. This self-healing test harness slashed manual regression time by >70 % and provides continuous coverage even as UI elements change. I also spearheaded “AI Brain,” a modular content-generation platform whose agent chains now power several niche products for our clients: interactive How-To guide creation, a long-form podcast generator, role-play skill assessments, and meeting-minute summarisation While further details are covered by NDA, these projects showcase my ability to weave together LLMs, autonomous agents, and traditional software engineering to create production-grade AI products that solve real business problems.

During my internship with GNX Group I engineered “AetherAgent,” a multi-agent, compliance-aware web-scraping bot. Built with LangChain LangGraph and the Stagehand browser-automation toolkit, the system orchestrates four specialised agents—Navigation, Action, Observation, and Compliance—to autonomously traverse JavaScript-heavy sites, extract high-value data, and validate results in real time. I embedded anti-detection tactics such as proxy rotation, human-like mouse-movement randomisation, dynamic typing cadences and automated CAPTCHA solving, while hard-coding legal safeguards that respect robots.txt rules and data-privacy statutes. The framework’s self-optimisation loop now adapts to layout changes without human intervention, reducing maintenance effort and accelerating data-collection throughput by several-fold.

Drove strategic initiatives to enhance operational efficiency and product competitiveness through AI integration and portfolio optimization. Worked closely with executive leadership (CEO, CFO, COO) and cross-functional teams to align insights with business goals. • Assessed the impact of generative AI on banking operations, identifying opportunities to streamline internal processes and improve customer experience. • Conducted market research and competitive analysis, delivering actionable recommendations to support strategic decision-making. • Proposed optimizations for deposit products, contributing to improved profitability and market positioning. • Presented findings to senior leadership, with several solutions currently under consideration for implementation. Enabled data-driven decision-making and supported innovation efforts aligned with Arab Bank’s long-term growth strategy.

As the lead Data Engineer for a Tier-1 banking client, I architected and maintained large-scale data-integration pipelines that moved 10 TB+ of transactional data each day across Netezza, Hadoop and on-prem SQL Server ecosystems. My remit covered end-to-end ETL design, performance tuning and production support, with a constant focus on shortening latency for downstream analytics. Impact highlights - Optimised ETL packages in Netezza, SSIS and SSRS, cutting average load time and unifying P2P, CRM and ERP feeds into a single enterprise data-warehouse. -Authored & regression-tested 500+ complex SQL scripts for Oracle E-Business Suite integrations (LO→LI, LI→L3), driving error-free nightly loads and improving financial-reporting SLA compliance - Developed & productionised SSIS/SSRS packages and ETL jobs that drive P2P analytics - Performed deep-dive root-cause analyses on data mismatches; remediations removed > 50 bottlenecks and reduced repeat incidents by 70 %.

Rotated into a dedicated QA analyst role for a major Canadian retail bank, validating end-to-end data and ETL pipelines that underpin credit-risk dashboards and reconciliations. Worked closely with data engineers and product owners to embed quality gates early in the SDLC. Key achievements & technologies: - Designed comprehensive test plans covering regression, integration and smoke testing for SSIS, Informatica and Netezza workflows processing 500 M+ daily records. - Automated data-set-to-dataset reconciliation scripts (Python, SQL) that flagged anomalies in seconds, cutting manual QA effort by 70 %. - Performed root-cause investigation on data mismatches between source (Core Banking System) and target marts, driving permanent fixes and preventing SLA breaches. - Logged, prioritised and closed 200+ defects through JIRA; facilitated daily stand-ups and sprint retros to ensure continuous improvement. - Collaborated with DevOps team to integrate data quality checks

Deans Lister