Omer Zulfiqar

The Automation Takeover: Are Software Engineers Becoming Obsolete? The speed at which LLMs have dominated the industry creates evidence of substantial change, including the evolving role of the software developer. from a 10,000-foot view, what is this signaling for what’s to come for jobs across sectors? LLMs are not just tools for generating text; they are becoming increasingly adept at understanding context, interpreting nuance, and engaging in meaningful conversations. This expansion of capabilities raises profound questions about the future role of software engineers. Are we entering an era where LLMs will take on more responsibilities traditionally handled by human developers? How will this impact the skill set required in the tech industry? Hessie Jones Jones explores the implications of LLMs' rise and the potential transformations awaiting the software engineering landscape in her thought-provoking piece. Check the first comment below. #ai #chatgpt #automation #softwareengineering #forbes #innovation

3

1 Comment

AI coding is a real thing but there's a few lessons learned: 1) If you don't know software, you should try it. It's amazingly useful 2) Anything of any degree of complexity makes it painfully obvious why Google's code standards do not permit try:except blocks. If an LLM's code doesn't work, it'll generally just say to wrap it in a try:except block. This works for small things, but can cause massive issues if you're trying to build something bigger 3) Linus Torvalds is famous for saying (paraphrased) "mediocre engineers care about their code, exceptional engineers care about their data structure". LLM generated code doesn't care about data structure and will often lead you astray. Fantastic tools -- not the end game yet. #ai

27

7 Comments

Examining AI related predictions from many (large/established) companies it is evident that the world of software & automation will be rebuilt with LLM at its core. Most of the dominant frameworks for developers to help set up pipeline, build agents, etc would be either open source or offerings from LLM providers/hyperscalers. If your company is not one of the above, you must have an extremely crisp, narrowly defined offering that shows how it accrues value for your prospects in the new world of LLM. #llm #agentiic #saas #openai #anthropic #devtools #langchain #claude #googlegemini

5

2 Comments

What should we be aware of when using LLMs? Google just released Gemma 2 to developers and the community: https://lnkd.in/gBTXZaZ2 This is Google’s next version of their open-weight LLM (different from Gemini their closed LLM). There have been great posts already about what Google has done differently with this model, but I wanted to highlight another part of this release. In the paper (https://lnkd.in/gYvS-Q_w) Google published along with Gemma 2, it covers the technical design of Gemma 2, but it also addresses 4 problems LLMs still face. 1. Catastrophic forgetting With an imbalance between the amount of data used for pre-training LLMs and fine-tuning LLMs, there are common cases where fine-tuning loses aspects from the initial data and "forgets" it in their responses. 2. Excessive verbosity When rewarding models for better structure in their answers to questions, this tends to deviate the entire LLM when having casual conversations. 3. Refusing to answer A problem that stems from the question: is no information better than incorrect information? 4. Exploitation *Note: this is actually from a different paper Google references: https://lnkd.in/gRigUZpt Policies dictate how LLMs learn using a reward system. But these systems, even if “perfect,” can be exploited using adversarial means. When we know how LLMs are designed to answer, we can find ways to exploit them. - I believe with more intelligent data and smaller language models, we can begin to address these issues!

7

Pathways to Tech Week #2 at  Headstarter: Being surrounded by people who are building amazing software and companies is a truly inspiring experience. My favorite speakers were: • Tariq Ahmed: First Engineer at AppLovin ($30Billion IPO) • Julian Alvarez 🚀: CEO of Wisdolia • Hamza Ali: CEO of Olostep  Here's what I have learned from them: 𝟭- "𝗙𝗮𝗶𝗹𝘂𝗿𝗲 𝗮𝗻𝗱 𝗽𝗶𝘃𝗼𝘁" Tariq, one of the first engineers at AppLovin, shared that AppLovin tried several things in their early stages that failed. However, the company had incredible success when it pivoted to mobile advertising; They didn't have expertise in the field. Today if you've experienced in-app advertising, AppLovin was likely behind it. 𝟮- "𝗦𝘁𝘂𝗱𝘆 𝘆𝗼𝘂𝗿 𝘂𝘀𝗲𝗿𝘀, 𝘆𝗼𝘂'𝗹𝗹 𝗯𝗲 𝘀𝗵𝗼𝗰𝗸𝗲𝗱 𝗮𝘁 𝗵𝗼𝘄 𝗺𝘂𝗰𝗵 𝘆𝗼𝘂 𝗰𝗮𝗻 𝗹𝗲𝗮𝗿𝗻." Julian runs Wisdolia a startup aimed at helping college students actively study. By simply asking what types of students they were. His team found that medical students were the ones the ones who converted the most into paying customers. It seemed evident because they were the most overwhelmed by the amount of information they had cram. But the evidence in their face was a lightbulb moment for Wisdolia. 𝟯- "𝗦𝗰𝗿𝗮𝗽𝗽𝗶𝗻𝗲𝘀𝘀 𝗮𝗻𝗱 𝗶𝗻𝗶𝘁𝗶𝗮𝘁𝗶𝘃𝗲 𝘄𝗲'𝗹𝗹 𝗴𝗲𝘁 𝘆𝗼𝘂 𝗳𝗮𝗿." Hamza, CEO of Olostep, describes his company as "a company that enables AI agents to browse the web just like humans". I was fascinated when I heard that, and as the Olostep team acquires more customers they value scrappy and people that initiative on board. Those qualities set engineers apart when building successful products.

11

1 Comment

I'm finding and summarizing interesting AI research papers every day so you don't have to trawl through them all. Today's paper is titled "Utilizing ChatGPT in a Data Structures and Algorithms Course: A Teaching Assistant's Perspective" by Pooriya Jamie, Reyhaneh Hajihashemi, and Sharareh Alipour. This paper explores the integration of ChatGPT into a Data Structures and Algorithms (DSA) course, emphasizing the role of teaching assistants (TAs) in guiding and supervising this integration. The research reveals that combining ChatGPT with structured prompts and active TA guidance can significantly enhance students' understanding of complex algorithmic concepts, promote engagement, and improve academic performance. Nevertheless, challenges such as maintaining academic integrity and addressing the limitations of large language models (LLMs) in solving complex problems remain pertinent. Here are some key findings from the study: 1. Enhanced Learning Outcomes: The study found that students using ChatGPT under TA supervision consistently outperformed those relying on traditional TA guidance only. Structured prompts and TA interactions were crucial in helping students engage with and understand the material deeply. 2. Role Evolution of TAs: The integration of ChatGPT has expanded the traditional roles of TAs. They now oversee AI tools, aiding in debugging, generating practice problems, and ensuring that AI outputs are educationally beneficial. 3. Model Comparison: Different versions of ChatGPT, namely 4o and o1, were employed to complement each other. ChatGPT-4o dealt with routine tasks, while ChatGPT o1, with advanced reasoning capabilities, assisted in tackling more complex problems, thus creating a dynamic learning environment. 4. Challenges and Limitations: Despite the benefits, challenges such as ChatGPT’s struggle with visual representations and highly complex algorithmic reasoning were highlighted. The study emphasizes the necessity of TA guidance to maximize the model's advantages while mitigating its drawbacks. 5. Scalability and Educational Impact: The hybrid approach of combining AI with human guidance makes the integration scalable for larger classes and reinforces the educational impact by reducing reliance on AI-generated content. The paper suggests that while LLMs like ChatGPT present promising educational tools, their effective implementation requires a balanced synergy between AI and human oversight. You can catch the full breakdown here: Here: https://lnkd.in/eHWHS3qr You can catch the full and original research paper here: Original Paper: https://lnkd.in/eSXmgqQr you're looking to improve your AI prompting skills, check out our free Advanced Prompt Engineering course: https://lnkd.in/ecB-XxY7 Follow for daily AI research paper breakdowns

1

Large language models are wonderful at ingesting large amounts of content & summarizing. Uploading an academic paper and I can pester it with an infinite list of questions & it will respond with equally infinite patience. In comparing the two most recent Microsoft earnings calls, Claude highlighted: - faster than expected Azure growth (29% vs 27%) - AI contributing 8% of Azure revenue up from 7% - higher CapEx spending & greater capacity constraints for data centers - much better commercial bookings growth : 17% vs 13% Excellent analysis. Benn Stancil described LLMs as great averagers of information. But what if I don’t want the average? If I seek the data point two standard deviations out? If I’m channeling my inner Anthony Bourdain & I’m seeking the fermented shark or cobra heart within my query? At Google, we ranked web pages & ads with many signals. An engineer taught me that of them found rarity of a word across a set of documents. For example, if there’s a document containing 10 instances of the word ghotiacross a collection of documents, where the average document has 0 ghotis, that document is likely the best to answer a search about fish. I haven’t found a way to goad an LLM to produce the rare result. Why did one user decide to use a particular piece of software when 10 others did not? Maybe I haven’t learned how to prompt an LLM well. Ideogram launched a feature called Magic Prompt that expands a basic prompt into richer instructions that elicit better results, marrying the language of a user & a computer. A user might write : A boy with a dog in a park. Magic Prompt replies : A heartwarming scene of a young boy playing with his happy, wagging dog in a lush, green park. The boy has a playful smile on his face while he tosses a bright yellow ball for his furry friend to fetch. The park is filled with beautiful flowers and trees, creating a serene atmosphere for this bonding moment between boy and dog. Magic Prompt surprised me in three ways. First, unlike Google search where terse queries tend to work very well, LLMs relish verbosity. Less Hemingway & more Faulkner. Second, maybe there’s a prompt for rarity. If you know what it is, please share with the rarity shibboleth. Third, I wonder what other prompt techniques are out there that have yet to be discovered to wrest the best from an LLM. This technique is called TF-IDF. It means Term Frequency Inverse Document Frequency.

22

6 Comments

Excited to see the launch of Andrej Karpathy's new venture’s first product, an AI course called LLM101n! This initiative leverages recent advancements in generative AI to create AI teaching assistants that guide students through course materials. It's a fantastic idea with a lot of potential. However, I'm curious about the business model. Given that this technology is achievable with RAG (Retrieval-Augmented Generation) and various LLMs, including open-source models, I have some doubts about the monetization potential. What are your thoughts on this? For more details, check out the article. #AI #generativeAI #newbusinessmodels #innovation #education

4

1 Comment

How I'm building AI companies. Step 1: Solve a problem for an industry that has a crap ton of money ✅ Real Estate is old and needs tech help, so for example my staging AI (Supastage - launching soon) could kick ass.

6

Product evals are the most under-discussed topics when it comes to building LLM products that *actually* produce results. LLMs are probabilistic by nature which means that building products around them is 10x more difficult than building traditional software. I believe that the best products will be built by teams who understand and create the best evaluation frameworks for their products. Article link in the comment section👇

23

6 Comments

Many AI startups that raise large $10m-$20m seed VC investments maybe better off by securing cloud computing infrastructure and/or hardware, instead of giving up a significant portion of their equity. Considering that most of the funds raised goes toward the infrastructure necessary for processing large AI/ML workloads. Startups accelerator programs by NVIDIA, Amazon and other financing options for hardware are worth considering. #startups #ai

35

🧠 Want LLM for phone call? Check out Moshi! Built by Kyutai, this experimental model is all about real-time chat - it can voice-act alongside its conversation partner. This kind of model will change every phone calls are made! Explore the platform here: https://moshi.chat/ Let's hear your thoughts! 💬 #moshi #ai

3

Some investors are still dismissing most AI startups as "GPT wrappers" -- i.e. they think companies that are using OpenAI or Anthropic will quickly get replaced or have no moat, since every company has access to these models. This is a flawed mental model. You can easily see the flaw if you think of OpenAI and Anthropic as providing intelligence, just as humans provide intelligence. You would never dismiss "a company that hires people" as "a human wrapper" and claim there's no moat because anyone can hire people. Instead, you have to think more deeply about accumulating advantages, network effects, product, expansion opportunities, etc. There are almost no YC startups that are literally just building a thin wrapper on OpenAI or Anthropic. They are all building real products, understanding customer problems, developing features that anchor them deeper into workflows, and solving problems that can't just be solved by using ChatGPT. A better general way of thinking about AI startups is The Black Hole Model, which I've talked about here before.

141

28 Comments

I've always believed that LLMs represent more of a UX revolution than just a shift in computing platform.    As more LLMs move from PoC to production, it's critical for companies to determine whether real users are getting value from the LLMs.   This means that a key part of the infra stack will be understanding user behavior at scale and tuning models to user preferences, rather than just using LLMs-as-a judge to evaluate performance.   I wrote a blog post about a user-centric framework that has successfully helped companies launch, evaluate, and improve LLMs at scale. Please check it out and let me know what you think. Link in the comments.

59

3 Comments

Agents and agentic platforms will pave the way for the future suffered delivery. There’s so much to be researched and experimented with… and, not to forget the questions of quality, security, and reliability. Those are the biggest challenges that will need to solve… And that isn’t even including the concept of human trust in what’s been created. Would you want to get onto an airplane knowing that the auto pilot software had been created by Devin? Groundbreaking potential! With challenges, limitations for us to address! #ArchAITecture #GenAI #SoftwareEngineering #AgenticPlatforms #DevSecOps

4

2 Comments

Quantization is the key to reduce cost. Through quantization, I've managed to reduce Llama 3.1's size while maintaining most of its capabilities. You get a pretty decent performance but with less cost. You need to explore quantization if you're working with LLMs.

3

Riding the AI wave by studying LLM theory or digging deep in ML models is a valid path, but IMO not the best. AI/ML/LLM nowadays can be used as-a-service (AaS) with relative ease, low costs and great availability of options. An optimal way to ride the wave, in my opinion, is to leverage these tools to build applications that solve real world problems. Basically, there are/will be: - Very few people doing ML theory and developing models and algorithms; - Several well-paid, high-skilled engineers building APIs and tools (often on/for the Cloud) that enable full-stack devs to leverage AI; - Many full-stack devs solving real world problems with AI through APIs (often on/from the Cloud). Agree?

12

1 Comment

Excited to welcome Lume (YC W23) to the KV portfolio! Lume uses AI to automate data mapping, empowering customer success & engineering teams to effortlessly understand, map, and validate data.

63

2 Comments

Slack, Instagram, and Zillow are worth a combined $167.4 billion. All of these products originated from hackathons. Before building FERMÀT, I spent 7+ years doing engineering at LiveRamp. And we held 4 hackathons a year. Here are the 3 most valuable lessons I learned: 1) Limited scope = Limited impact 10 full days. That's how long each of our quarterly hackathons lasted. This was how long we needed to build features that could have a 10X impact. My team is currently running 3-day hackathons twice a year and is actively planning to expand this. 2) They boost team morale Many of the best engineers I have worked with rally around hackathons. It's like a sporting event, they get super excited, work incredibly hard with laser focus, and work together to achieve something amazing. 3) They deserve to be part of your product roadmap. Some of the best features in the world owe their origins to hackathons. It’s your leadership team's job to identify the best projects so they are incorporated into your product roadmap. This ensures that your team treats hackathons as an opportunity to have a real impact on your product, as opposed to a waste of valuable time. What do you think—are hackathons a good investment in 2024?

33

5 Comments