AI

ENGINEERING

HACKATHON POWERED
BY

RULE BOOK &

PROBLEM STATEMENT

for more information visit

poridhi.io/hackathon
 AI
ENGINEERING

HACKATHON POWERED
BY

Artificial Intelligence is revolutionizing software engineering

and reshaping the global tech ecosystem. From automation to

data-driven decision-making and intelligent system

development, AI is unlocking new opportunities for

developers. However, integrating AI into software requires a

new kind of engineering mindset. Simply making API calls to an

LLM isn’t enough—AI’s power comes with unpredictability,

demanding innovative engineering approaches.

Many engineers in the industry are already working with AI or

will soon need to. To ensure that both professionals and

students stay ahead in this evolving landscape, Poridhi and

Brain Station 23 are bringing you AI ENGINEERING

HACKATHON, a platform to compete, learn, and innovate in AI

system building. Get ready to push the boundaries of AI

engineering!
 AI
ENGINEERING

HACKATHON POWERED
BY

NAME

Poridhi presents AI Engineering Hackathon

Powered by Brain Station 23

DATE

26 April, 2025

VENUE

BRAIN STATION 23
12TH FLOOR, 1 BIR UTTAM AK
KHANDAKAR ROAD, DHAKA 1212

TIME

8.00 AM- 9.15 PM

 AI
ENGINEERING

HACKATHON POWERED
BY

PRIZE MONEY

CHAMPION TEAM

2,50,000 BDT
FIRST RUNNERS -UP TEAM
DATE
50,000 BDT
SECOND RUNNERS -UP TEAM
30,000 BDT
SPECIAL MENTION TEAM

20,000 BDT
WHAT WILL YOU GET

Participation Certificate for all the participants.

Multiple training sessions for the hackathon

preparation.

Access to 20+ Exclusive Hands-on Labs from

Poridhi

Breakfast, dinner and evening snacks on the event.

 AI
ENGINEERING

HACKATHONPOWERED
BY

PROGRAM OUTLINE

TIME PARTICULARS

7.30 AM - 8.00 AM Attendance & Registration

8.00 AM - 8.15 AM Inauguration Program

briefing & hackathon

8.15 AM - 8.30 AM
starts

9.00 AM - 9.30 AM breakfast & refreshment

12.00 pM - 2.00 pM prayer & lunch break

4.30 pM - 5.00 pM evening snacks

7.00 pM - 8.30 pM presentation

8.30 pM - 8.45 pM keynote speech

8.45 pM - 9.15 pM prize giving ceremony

 AI
ENGINEERING

HACKATHON POWERED
BY

CODE OF CONDUCT

Respectful Communication: Participants are expected to

communicate respectfully with each other, including avoiding

the use of profanity or derogatory language. We ask that

DATE
everyone refrains from making personal attacks or engaging in

harassment, discrimination, or any other behavior that could

make others feel uncomfortable or unwelcome.

Inclusive Attitude: We encourage diversity and inclusivity in

all forms, and we expect participants to maintain a welcoming

and supportive attitude towards everyone, regardless of their

gender, race, ethnicity, nationality, age, or religion.

Professionalism: Participants are expected to behave

professionally and act in good faith throughout the event. This

includes adhering to the rules and guidelines of the hackathon,

working collaboratively with others, and respecting the

intellectual property of others.

 AI
ENGINEERING

HACKATHON POWERED
BY

CODE OF CONDUCT

Reporting Concerns: If a participant witnesses or

experiences behavior that violates the code of conduct, they

are encouraged to report it to the hackathon organizers

DATE
immediately. We take all reports seriously and will investigate

any concerns fairly and impartially. The organizer committee

has the full right to disqualify any participant and void the

championship if they are found violating any of the rules or if

their conduct is deemed inappropriate.

By following this code of conduct, we can create an

environment that fosters creativity, collaboration, and

innovation. Violations of the code of conduct may result in

consequences such as warnings, temporary or permanent

bans, or legal action if necessary.

 AI
ENGINEERING

HACKATHON POWERED
BY

PROBLEM STATEMENT
DATE
The “AI Engineering Hackathon 2025 powered by Brain

Station 23” challenges teams to push the boundaries of

intent-based product search, where engineering excellence

meets real-world AI scalability. Participants will tackle the

complexities of AI-driven search, balancing precision,

performance, and scalability while ensuring a robust and

efficient deployment pipeline.

From fine-tuning embedding models to architecting resilient

search infrastructures, teams will be judged on their ability to

navigate key AI engineering principles.

KEY ENGINEERING CHALLANGES

✅ Scalability & Deployment: Architect a system that

maintains low latency under high traffic.

✅ Model Optimization: Fine-tune embeddings and optimize

ranking models for relevance.

✅ Guardrails & Safety: Implement hallucination prevention,

DATE
bias mitigation, and ethical AI principles.

✅ Monitoring & Observability: Integrate tracing, logging,

drift detection, and real-time monitoring.

✅ Architectural Decisions: Evaluate RAG vs. agentic

workflows or engineer a novel hybrid solution.

✅ Cost & Latency Optimization: Reduce compute overhead

while improving inference efficiency.

✅ Dataset Engineering: Design robust data pipelines for

fine-tuning, indexing, and retrieval.

✅ Infrastructure & Security: Build scalable, resilient

architectures while implementing API security, rate limiting,

and access controls.

✅ Continuous Improvement: Implement automated

feedback loops for self-improving search accuracy.

KEY ENGINEERING CHALLANGES

This hackathon is about tackling real-world AI engineering

challenges at scale. Can you design a high-performance,
secure, and scalable search system that truly understands
user intent?

🔹 How will you optimize your

DATEsearch pipeline for speed and
efficiency?
🔹 Will you choose RAG, agentic workflows, or invent a new
hybrid model?
🔹 Can you minimize inference costs while maximizing retrieval
accuracy?
🔹 How will you engineer thescalable data pipelines for fine-
tuning and continuous learning?
🔹 Can your system detect drift, monitor the real-time
performance, and ensure observability?
🔹 How will you deploy a security-first architecture that
protects data and ensures API resilience?

The winning teams will prove their engineering

expertise, strategic decision-making, and ability to
scale AI systems in production.