esolutions.tech

27/05/2026

25 Years. Hundreds of projects. One eSolutions.

🚀 We’re officially marking a milestone that few tech companies reach: 25 years in business. While the tools we use have changed completely since our first day, our core approach remains the same:
🔹 Constant evolution: We’ve grown alongside the digital age, evolving our skills to ensure our partners stay ahead of the curve.
🔹 Results over hype: We’ve never been about the latest buzzword. We’re about building solutions that work, scale, and last.
🔹 Human foundation: Beyond the code and the systems, eSolutions is built on the people who work here and the clients who trust us. Those relationships are the real achievement.

It’s been an incredible 25-year run, and we’re just getting started. A huge thank you to everyone who has been part of the eSolutions story!

hashtag hashtag hashtag hashtag

26/05/2026

We’re happy to see our colleague Roberto Comsa speaking at the upcoming Big Data Meetup in Bucharest.
At eSolutions.tech, we always appreciate opportunities to share knowledge, exchange ideas with others in the industry, and be part of the local tech community. It’s great to see people openly talking about real challenges, lessons learned, and the work happening behind the scenes.
If you’re into data and engineering, it’s definitely worth joining the conversation and meeting others in the field.

Why does data validation matter so much in CDC streaming pipelines? Roberto Comsa, Data Engineer at eSolutions.tech, will explore this scenario through real production systems.

Working with NiFi, Kafka, Spark, Delta Lake, and MinIO, he’ll walk through how CDC data flows from ingestion to consumption and how small issues in streaming pipelines can quietly impact data quality over time. The focus will be on practical validation strategies, early detection of drift, and building trust in continuously moving data.

📍 June 9 | ING Hubs Romania

🎟️ Free entry

👉 Looking to join? More details and RSVP here: https://www.meetup.com/bucharest-big-data-meetup/events/314642829

19/05/2026

With Alteus Medical, we’re building practical AI tools that reduce clinical admin work and help doctors focus more on patients, from consultation support to automated documentation and full traceability.

Excited to share what’s already possible in healthcare AI today.

13/05/2026

Yesterday at the Women in Tech Global Conference® 2026, our colleague Corina Staicu, CPO, spoke about a topic many organizations are navigating today: how AI can work with legacy systems, not just replace them. In her session, “AI Over Legacy Systems,” she challenged the assumption that older systems automatically need to be rebuilt from scratch.

One of our favorite lines from the talk was: “Legacy doesn’t mean old. It means hard to change.” Instead of ripping everything out and starting over, Corina explored the idea of AI overlays: adding intelligence on top of existing systems, while keeping the parts that already work.

The discussion covered topics such as:
• The impact of AI on software delivery speed
• Why large-scale replacement projects still struggle
• How conversational interfaces are reshaping enterprise UX
• The importance of governance and observability from the beginning
• How organizations can start with smaller, incremental changes

Thank you to the Women in Tech team for the invitation and the conversation.

07/05/2026

A practical, clinician-centered approach to AI in healthcare. Excited to introduce Alteus Medical, our AI-powered clinical copilot built to help doctors spend less time on documentation and more time on what matters most: patient care.

05/05/2026

AI becomes enterprise-critical when it moves from experimentation to being embedded in real-world workflows: supporting automation, decision-making, and day-to-day operations. At that stage, governance matters: risk management, transparency, data privacy, and regulatory compliance become part of how solutions are designed and deployed.

We outlined our approach to AI strategy, readiness, integration, RAG, AI agents/multi-agent systems, and AI governance here: https://www.esolutions.tech/ai-services

30/04/2026

In retail environments, data silos tend to emerge as systems optimise for store operations, e-commerce performance, supply chain visibility, or loyalty programs in isolation.

Each platform supports a legitimate objective. Over time, however, customer records fragment, inventory views diverge, and promotional analytics reflect partial context. Decision-making then depends on stitched reports rather than shared foundations.

In practice, silos expose coordination boundaries across channels and functions. We explored how to move toward a unified, real-time data architecture to regain a 360-degree customer view and scale effectively here: https://www.esolutions.tech/data-silos-in-retail

23/04/2026

While teams are eager to experiment, the AI sprawl often creates significant governance risks. We see this gap often: you have the tools, but lack the centralised oversight to manage permissions, data context, and costs safely.

That’s exactly why we built Alteus.ai. It provides the administrative backbone for the modern AI stack, turning fragmented experiments into a unified, enterprise-grade environment.

Everyone is excited about AI, but most companies miss the real challenge: AI sprawl. Different teams experiment with different tools. Different models. Different prompts. Different data sources.

Soon, you have 20 AI experiments, but zero governance. This is where many organizations get stuck. What companies usually forget about AI adoption is that AI should be under control.
For example:
• Who can access which AI models?
• Which internal documents can the AI use?
• Which departments can deploy AI agents?
• How do you monitor usage and cost?

Without a centralized platform, AI becomes chaotic. That’s why modern AI stacks are starting to include an AI administration console. A place where organizations can:
• manage user permissions
• control access to models
• define data contexts per department
• monitor usage and AI activity

Platforms like Alteus.ai take this approach by unifying AI models, company knowledge, and internal tools into one controlled environment.

How do you manage AI at scale?

21/04/2026

We see the potential everywhere: AI assistants that can assess symptoms, summarise patient histories, and flag high-risk cases. But for these tools to move from Proof of Concept to Point of Care, they must cross the bridge of clinical adoption.

However, to move from experimental to operational, three things must be true:

1️⃣ Reliability: data quality and explainable outputs are non-negotiable.
2️⃣ Workflow: AI should fit into existing EHR systems to reduce (not increase) the administrative load.
3️⃣ Accountability: clear human oversight ensures that AI is used to augment (and never replace) the clinician’s expertise.

Read more on our approach to implementing AI in high-stakes healthcare environments: https://www.esolutions.tech/ai-in-healthcare

16/04/2026

In the rush to move to agentic workflows, companies may fall into a familiar trap: embedding system-specific logic directly into the application layer. The result? A spaghetti integration where changing a model or updating a database schema could break the entire pipeline.

The Model Context Protocol (MCP) offers a more resilient architectural path. By decoupling reasoning (the model) from ex*****on (the connectors), MCP allows for a modular AI stack:

🔹 Model agnosticism: possibility to swap LLMs without rewriting the integration logic.
🔹 Separation of concerns: the model focuses on intent, while MCP connectors handle the technical "how-to" of accessing APIs or internal databases.
🔹 Scalability: adding proprietary connectors as the environment evolves, without adding complexity to the core agent logic.

We dive deeper into why MCP is becoming the standard for enterprise-grade AI architecture here: https://www.esolutions.tech/mcp-enterprise-ai-integration